~The Delhi Sultanate

A Political and Military History

The Delhi Sultanate was the first Islamic state to be established in India. In broad-ranging and accessible

arrative, Peter Jackson traces the history of the Sultanate from its foundation in 1210 to its demise in around 1400 following the sack of Delhi by the Central Asian conqueror, Temur (Tamerlane). During the

thirteenth and fourteenth centuries, the Sultanate was the principal bastion of Islam in the subcontinent.

While the book focuses on military and political affairs, tracing the Sultanate's expansion, its resistance to

formidable Mongol invasions from the northwest and the administrative developments that underpinned

these exploits, it also explores the Sultans' relations with their non-Muslim subjects. As a comprehensive

treatment of the political history of this period, the book will make a significant contribution to the

literature on medieval Indo-Muslim history. Students of Islamic and South Asian history, and those with a

general interest in the region, will find it a valuable resource.

PETER JACKSON is Senior Lecturer in the Department of History at Keele University. He is editor of The

Cambridge History of Iran, volume 6 (1986), and translator and joint editor of The Mission of Friar

William of Rubruck (1990).

~Cambridge Studies in Islamic Civilization

Editorial board

DAVID MORGAN (general editor)

VIRGINIA AKSAN MICHAEL BRETT MICHAEL COOK PETER JACKSON TARIF KHALIDI ROY

MOTTAHEDEH BASIM MUSALLAM CHASE ROBINSON

Titles in the series '

STEFAN SPERL, Mannerism in Arabic poetry: a Structural Analysis of Selected Texts, 3rd Century AHI9th

Century AD-5th Century AH/1 lth Century AD 0 521 354854 PAUL E. WALKER, Early Philosophical Shiism: the Ismaili Neoplatonism of Abu Ya'qub al-Sijistani 0 521 441293 BOAZ SHOSHAN, Popular Culture

in Medieval Cairo 0 521 43209X STEPHEN FREDERIC DALE, Indian Merchants and Eurasian Trade, 1600-

1750, 0 521454603 AMY SINGER, Palestinian peasants and Ottoman Officials: Rural Administration

around Sixteenth-century Jerusalem, 0 521 452384 (hardback) 0 521 476798 (paperback) TARIF KHALIDI,

Arabic Historical Thought in the Classical Period, 0 521 465540 (hardback) 0 521 58938X (paperback)

LOUISE MARLOW, Hierarchy and Egalitarianism in Islamic Thought, 0 521 564301 JANE HATHAWAY, The

Politics of Households in Ottoman Egypt: the Rise of the Qazdaglis, 0 521 571103 THOMAS T. ALLSEN,

Commodity and Exchange in the Mongol Empire: A Cultural History of Islamic Textiles, 0 521 583012

DINA RIZK KHOURY, State and Provincial Society in the Ottoman Empire: Mosul, 1540-1834, 0 521 590604

THOMAS PHILIPP AND ULRICH HAARMANN (eds.), The Mamluks in Egyptian Politics and Society, 0 521

591155

~The Delhi Sultanate

A Political and Military History

PETER JACKSON

CAMBRIDGE

UNIVERSITY PRESS

~PUBLISHED BY THE PRESS SYNDICATE OF THE UNIVERSITY OF CAMBRIDGE The Pitt

uilding, Trumpington Street, Cambridge CB2 1RP, United Kingdom

CAMBRIDGE UNIVERSITY PRESS

The Edinburgh Building, Cambridge CB2 2RU, UK http://www.cup.cam.ac.uk 40 West 20th Street, New

York, NY 10011-4211, USA http://www.cup.org 10 Stamford Road, Oakleigh, Melbourne 3166, Australia

© The Cambridge University Press, 1999

This book is in copyright. Subject to statutory exception and to the provisions of relevant collective

licensing agreements, no reproduction of any part may take place without the written permission of

Cambridge University Press.

First published 1999.

Printed in the United Kingdom at the University Press, Cambridge

Typeset in Times 10/12pt CE

A catalogue record for this book is available from the British Library

Library of Congress Cataloguing in Publication data

Jackson, Peter

The Delhi Sultanate : a political and military history / Peter Jackson p. cm. - (Cambridge studies in Islamic civilization)

Includes bibliographical references and index.

ISBN 0 521 40477 0

1. Delhi (Sultanate) - History. I. Series.

2. DS459.J27 1998 954'.56023-dc21 98 30080 CIP

ISBN 0 521 40477 0 hardback

~For Rebecca

~Contents

List of maps page xi

Preface xiii

Note on transliteration xvi

List of abbreviations xvii

The Thirteenth Century 1

1 The background 3

2 From Ghurid province to Delhi Sultanate 24

3 Sultans and sources 44

4 Turks, Tajiks and Khalaj 61

5 The centre and the provinces 86

6 The Mongol threat 103

7 Raid, conquest and settlement 123

The Zenith of the Sultanate 149

8 Sultans, saints and sources 151

9 The Khalji and Tughluqid nobility 171

10 An age of conquest 193

11 The Chaghadayid invasions 217

12 The military, the economy and administrative reform 238

13 Stupor mundi: the reign of Muhammad b. Tughluq 255

14 The sultans and their Hindu subjects 278

15 Stasis and decline: Firuz Shah and his successors 296

Epilogue: c. 1400-1526 321

~x Contents

Appendices 326

I The term 'Turk' 326

II Qilich Khan Mas'ud b. cAla' al-Din Jani 327

III Qara'unas and Neguderis 328

IV ‘Ayn al-Mulk Multani and ‘Ayn al-Mulk Ibn Mahru 329

V The date of Ghiyath al-Din Tughluq Shah's death 330

VI The ancestry of Tughluq Shah II 332 Genealogical tables 333

Glossary 336

Select bibliography 339

Index 351

~Maps

1 The eastern Islamic world in 1206 page 23

2 The frontier with the Mongols 120

3A and 3B The war against Hindu powers in northern India 131 and 137

4 The conquest of Gujarat, Malwa and the south 212

5 The cities of Delhi 259

6 The Sultanate under the Tughluqids 297

XI

~preface

This book is designed to be a political and military history of the 'Greater' Delhi Sultanate, which after its

creation in 1210 lasted for almost two hundred years and for almost half that period functioned as the sole

bastion of Muslim power in the Indian subcontinent. The era from the sack of Delhi by the Central Asian

conqueror Temur (Timur-i Lang, 'the Lame'; Westernized as 'Tamerlane') in 801/1398 down to the Mughal conquest in 932/ 1526, during which the Sultanate was merely one of several competing Muslim kingdoms

in the north, is briefly covered in the Epilogue.

The source materials for the Delhi Sultanate - largely narrative in form and written in Persian, with

the addition of descriptions of India by external observers who wrote in Arabic - are markedly less

satisfactory than, for instance, either those available for the Mughal empire that followed it or those

composed in the contemporary Mamluk Sultanate of Egypt and Syria. Much of the general literature on this

period of Indian history has tended to adhere, in my view, far too closely to the arrangement in the narrative

sources, and accordingly the reader is all too often served up a barely digestible repast of seemingly

unconnected events.

I have divided the period into two phases, with the reign of ‘Ala' al-Din Khalji (695-715/1296-

1316) marking a watershed: his era witnessed the implementation of far-reaching administrative changes,

designed in large part to meet both an escalation in Mongol attacks and a more vigorous advance in Rajasthan and the south. Each of the two sections is introduced by a chapter on the sources, and the view

they purvey of the sultans; but otherwise, within each section I have tried to approach the task themati-

cally, giving prominence to the formation of the aristocracy, to administrative control and to the perennial

warfare against the Sultanate's enemies, whether independent Hindu powers or the Mongols of Afghanistan

and Central Asia. In chapters 12-13 and 15 an attempt has been made to bring Political and military affairs

into relation with economic developments, although it has to be said that material for the economic history

of the Sultanate is relatively meagre. Two chapters, focusing on the reigns of Muhammad bin Tughluq

(724-752/1324-51) and of Firuz Shah (752-790/

xin

~xiv Preface

1351-88), represent a departure from the framework I have adopted; but it seemed advisable to devote a

consolidated study to each of these problematic reigns. It is hoped that chapter 14, on the sultans' relations

with the subject Hindu population, fits naturally between them, given Muhammad's favour towards Hindus

and his successor's allegedly more rigorous attitudes.

This book has been some years in gestation, and in writing it I have accumulated many debts. It is

a pleasure to be able at last to acknowledge an award from the Leverhulme Trust which contributed

towards the cost of replacement teaching for two terms in 1990-1, and the generosity of Keele University

both in meeting the balance of those costs and in granting me a research award for a further term and

funding research expenses. Thanks are also due to my medievalist colleagues in the History department for

closing ranks when I was on sabbatical leave. I have benefited greatly from the assistance of the inter-library loans section of Keele University Library, and from the facilities offered by the Cambridge

University Library, the Oriental Room of the Bodleian Library and the Indian Institute in Oxford, the John

Rylands University Library at Manchester, the India Office Library and the Oriental Students' Room of the

British Library (now amalgamated), the Library of the School of Oriental and African Studies in the

University of London, the Bibliotheque Nationale in Paris, and the Library of the Rijksuniversiteit Leiden.

The forbearance of the Librarian of the Royal Asiatic Society towards a notoriously long-term borrower is

also deeply appreciated. I am grateful to the relevant Turkish authorities for permission to consult the

manuscript collections in the Suleymaniye and Nuruosmaniye Libraries and the Topkapi Sarayi Muzesi in

Istanbul. Dr Renato Traini, librarian at the Biblioteca dell'Accademia Nazionale dei Lincei e Corsiniana in

Rome, promptly and courteously supplied me with photocopies of the relevant folios of the manuscript

Caetani 21 of al-Safadi's al-Wafi bi'l-Wafayat. The Bodleian Library, the British Library and the National Archives of India have also kindly provided me with microfilms of certain manuscripts in their collections.

A number of scholars contributed towards the production of this book. Some years ago, Mr Simon

Digby generously lent me a photocopy of most of the manuscript of the first recension of Barani's Ta'rikh-i

Firuz-Shahi in his private collection, which has proved invaluable, and more recently gave me permission

to use a text in which he has collated the portion of this manuscript covering the reign of Muhammad b.

Tughluq with the relevant section of that in the Bodleian Library. In India in 1991, Dr Akbar Ali Khan

Arshizade, Officiating Director of the Raza Library at Rampur, extended to my wife and myself a

hospitality we still remember with warm gratitude. We had good reason, too, to value the assistance of

Vikram, our driver, and Toni, our guide in the old city of Delhi. For the production of the maps I am

indebted to my colleague Andrew Lawrence, of the

~Preface xv

Cartographic unit in the Department of Environmental Social Sciences at Keele. At the Cambridge

University Press, Marigold Acland has proved an extremely patient and good-natured editor.

It will be obvious in the following pages how much I have profited from the work of other

scholars who have made the eastern Islamic world, and in particular Muslim India, very much more their

field than I have myself. Dr Peter Hardy and Professor Edmund Bosworth, who jointly examined my PhD

thesis in 1976, have continued to sustain me with their friendship, interest and hospitality. I have gained

also from the opportunity to meet and argue about the Delhi Sultanate with Dr Khurram Qadir, of the

Bahauddin Zakariya University at Multan. Naturally, I enjoy undivided credit for any errors that have crept

into the book. My greatest debt is acknowledged, inadequately, in the dedication. Despite the heavy demands of

her own career, my wife has never failed to offer encouragement and moral support to an author who at

times appeared to be teetering on the edge of insanity. Without her this book could not have been written.

~Note on transliteration

For the transliteration of Arabic and Persian, I have used the system adopted in the Encyclopaedia of Islam,

except that ch is employed instead of c, j for dj, and q for k. For the sake of uniformity, Persian names and

terms derived from Arabic are spelled as if they were Arabic: thus Muhaddith rather than Muhaddis,

dhimma for zimma, hadrat for hazrat, and waqf in place of vaqf. The Persian idafa has been rendered

throughout as -[y]i. For Turkish and Mongol proper names and terms, I have followed the UNESCO system, as employed in J. A. Boyle, The successors of Genghis Khan (New York, 1971). The tentative

reconstruction of a proper name is indicated by an asterisk, as in *Altunapa or *Tartaq. Precise readings, as

found in manuscripts or printed texts, are reproduced in capitals, with X standing for kh, T for gh, C for ch,

S for sh, Z for zh, ' for hamza, and the long vowels represented by A, W and Y (a 'tooth' without diacritical

poin ts appears as a dot).

Indian names present a greater problem, and here I have undoubtedly been guilty of inconsistency.

The names of those places that found their way into standard Islamic geographical lore are given in Arabic-

Persian form, e.g. Qinnawj and Bada'un in place of Kanauj and Budaon; but otherwise a hybrid (if

hopefully recognizable) form has been employed, e.g. Kol, Chanderi, Erach, rather than Kul, Chandiri,

Irach. Where a European spelling has become established, however, as with Delhi and Lahore, I have given

the Persian-Arabic form (Dilti, Dihli; Lahawr) alongside it at first encounter, thereafter adhering to the

form in common use.

xvi

~Abbreviations

Periodicals and reference works

AEMA Archivum Eurasiae Medii Aevi

AOH A cta Orientalia Academiae Scientiarum Hungaricae

ARIE Archaeological Survey of India. Annual Report on Indian Epigraphy

BEO Bulletin d'Etudes Orientates de l'nstitut Francais de Damas

BI Bibliotheca Indica

BL British Library

BN Bibliotheque Nationale, Paris

BSOfAJS Bulletin of the School of Oriental [and African] Studies, University of London

CAJ Central Asiatic Journal

CCIM H. Nelson Wright (ed.), Catalogue of the coins of the Indian Museum, Calcutta

CMSD H. Nelson Wright (ed.), The coinage and metrology of the Sultans of Dehli

DGUP District gazetteers of the United Provinces of Agra and Oudh (Allahabad, 1903-22, 48 vols.)

ED Sir Henry Elliot, A history of India as told by its ownhistorians, ed. J. Dowson (London, 1867-77,

8 vols.) El Epigraphia Indica

EIAPS Epigraphia Indica. Arabic and Persian Supplement

EIM Epigraphia Indo-Moslemica

Enc.Ir . E. Yarshater (ed.), Encyclopaedia Iranica (London and Costa Mesa, California, 1982- in

progress)

Enc.Isl.2 Ch. Pellat et al. (eds.), The encyclopaedia of Islam, new edn (Leiden, 1954- in progress) GMS

Gibb Memorial Series

HI Hamdard Islamicus

HJAS Harvard Journal of Asiatic Studies

xvii

~xviii Abbreviations

HN M. Habib and K. A. Nizami (eds.), The Delhi Sultanat (A.D.1206-1526)

HS Hakluyt Society

I A Indian Antiquary

IC Islamic Culture

IESHR Indian Economic and Social History Review

IG W. S. Meyer et al. (eds.), The Imperial Gazetteer of India, new edn. (Oxford, 1907-9, 26 vols.)

IHQ Indian Historical Quarterly

IHR Indian Historical Review

IOL India Office Library, London

IO[N]S Israel Oriental [Notes and] Studies

Iran Iran. Journal of the British Institute of Persian Studies

IS Islamic Studies

IU Islamkundliche Untersuchungen

JA Journal Asiatique

JAH Journal of Asian History

JAOS Journal of the American Oriental Society

JAS[B] Journal of the Asiatic Society [of Bengal]

JASP Journal of the Asiatic Society of Pakistan

JB[O]RS Journal of the Bihar [and Orissa] Research Society

JCA Journal of Central Asia

JIH Journal of Indian History

JIS Journal of Islamic Studies

JNSI Journal of the Numismatic Society of India

JPHS Journal of the Pakistan Historical Society

JRAS Journal of the Royal Asiatic Society of Great Britain and Ireland

JSS Journal of Semitic Studies

JUPHS Journal of the United Provinces Historical Society

MA SI Memoirs of the Archaeological Survey of India

MIM Medieval India: a miscellany

MIQ Medieval India Quarterly

NIA New Indian Antiquary

PFEH Papers on Far Eastern History

PIHC Proceedings of the ... Indian History Congress [numeral refers to the number of the session]

PL C. A. Storey, Persian literature: a bio-bibliographical survey (London, 1927- in progress)

PPV Pamiatniki Pis'mennosti Vostoka

PSMI Proceedings of the Seminar on Medieval Inscriptions (6—8th Feb. 1970) (Aligarh, 1974)

PUJ Patna University Journal

~OGIA Deutsches Archaologisches Institut, Kairo. Quellen zur Geschichte des islamischen Agyptens

RCEA Et. Combe, J. Sauvaget and G. Wiet (eds.), Repertoire chronologique d'epigraphie arabe (Cairo,

1931- in progress)

RRL Rampur Raza Library

SK Suleymaniye Kutuphanesi, Istanbul

SOAS School of Oriental and African Studies, University of London

TMENP G. Doerfer, Turkische und mongolische Elemente im Neuper- sischen

TSM Topkapi Sarayi Muzesi, Istanbul

TVOIRAO Trudy Vostochnago Otdeleniia Imperatorskago Russkago Ar- kheologicheskago Obshchestva

WZKM Wiener Zeitschrift fur die Kunde des Morgenlandes

ZS Zentralasiatische Studien

Texts

AHG Ulughkhani, Zafar al- Walih, ed. Ross, An Arabic history of Gujarat

AH Fakhr-i Mudabbir, Adab al-Harb wa'l-Shaja'a Babur-Nama Babur, Bdbur-Nama

CN Kufi, Chach-Nama

DA Ghaznawi, Dastur al-Albab

DGK Amir Khusraw, Dibacha-yi Ghurrat al-Kamal

DR Amir Khusraw, Diwal Rdni-yi Khadir Khan

FFS Sultan Firuz Shah Tughluq, Futuhdt-i Flruz-Shahi

FG Yusuf-i Ahl, Fard'id-i Ghiyathi

FJ Barani, Fatawa-yi Jahdnddri

FS cIsami, Futuh al-Salatin

GK Amir Khusraw, Ghurrat al-Kamal

IA Ibn al-Athir, al-Kamilfi'l-Ta'rikh

IB Ibn Batttuta, Tuhfat al-Nuzzar

IM Ibn Mahru, Insha-yi Mahru

JH ‘Awfi, Jawami al-Hikdyat

JT Rashid al-Din Fadl-Allah, Jami al-Tawarikh

KF Amir Khusraw, Khazd'in al-Futuh

MA al-‘Umari, Masalik al-Absar

MF Amir Khusraw, Miftah al-Futuh

NS Amir Khusraw, Nuh Sipihr

QS Amir Khusraw, Qiran al-Sa’dayn

RI Amir Khusraw, Rasa'il al-I’jaz

SA Fakhr-i Mudabbir, Shajarat al-Ansab

SFS Anonymous, Sirat-i Firuz-Shahi

Siyar Kirmani (Amir Khwurd), Siyar al-Awliya'

~xx Abbreviations

SP Rashid al-Din Fadl-Allah, Shucab-i Panjgana

Taj Hasan-i Nizami, Taj al-Ma'athir

TFS Barani, Ta'rikh-i Firuz-Shahi

TFS1 Barani, Ta'rikh-i Firuz-Shahi, first recension

TJG Juwayni, Ta'rikh-i Jahan-Gusha

TMS Sirhindi, Ta'rikh-i Mubarak-Shahi

TN Juzjani, Tabaqat-i Nasiri

TS Amir Khusraw, Tuhfat al-Sighar

Tughluq-Nama Amir Khusraw, Tughluq-Nama

WH Amir Khusraw, Wasat al-Hayat

Shami, ZN Shami, Zafar-Nama

Yazdi, ZN Yazdi, Zafar-Nama

~PART I

The thirteenth century ~ ~CHAPTER 1

The background

Caliphs, amirs and sultans

The ghosts of two great Muslim conquerors haunted the rulers of the Delhi Sultanate. One was Mahmud of

Ghazna (d. 421/1030), whose campaigns had extended Islamic rule into the western Panjab. The other was

the Ghurid Sultan Mucizz al-Din Muhammad b. Sam, whose more recent victories over a number of Hindu

states had entrenched Muslim power in the north Gangetic plain, and whose murder in 602/1206 had first

propelled Muslim India on its own separate path, distinct from that taken by the lands west of the Indus.

Mahmud and Mu'izz al-DIn, each in his way, typified the warlords who had been carving out principalities

for themselves within the Islamic world since the ninth century. The universal Caliphate of the 'Abbasids

had steadily disintegrated, leaving them with only the titular headship of the orthodox (Sunni) Muslim

community. Some provinces had been lost to the heterodox Shifts. For almost three centuries (296-567/

909-1171) the cAbbasids were challenged by the Fatimid Imams representing the Isma'ili Shift sect. From

Egypt and Syria, these counter-caliphs deployed a network of agents and propagandists whose activities

extended even as far east as Sind, the region of the middle and lower Indus valley, reduced by the caliphal general Muhammad b. Qasim al-Thaqafi as early as 92/711. From 344/965 the Fatimid Imam's name was

mentioned in the prayers at Multan, and by the end of the century at Mansura.1 But in the majority of

caliphal territories power passed into the hands of semi-independent, hereditary governors. Such rulers,

who initially bore no title higher than amir (literally 'commander'), usually went through the formality of

obtaining a patent of authority (manshur), a robe (khil'at) and a sonorous

For a good introduction to the first centuries of Muslim rule in Sind, see Yohanan Friedmann, 'A

contribution to the early history of Islam in India', in M. Rosen-Ayalon (ed.), Studies in memory of Gas ton

Wiet (Jerusalem, 1977), 309-33; Derryl N. MacLean, Religion and society in Arab Sind (Leiden, 1989). On

Ismacil! activity, see S. M. Stern, 'Ismail! Propaganda and Fatimid rule in Sind', IC 23 (1949), 298-307; cAbbas al-Hamdani, The beginnings of the Isma'il! Da’wa in northern India (Cairo, 1956).

~4 The thirteenth century

title (laqab) from the ‘Abbasid Caliph, in return for inserting his name in the public Friday sermon (khutba)

and on the coinage (sikka) and, more notionally, remitting an annual tribute.

To bolster their dubious legitimacy, the provincial amirs had to act (or pose) as champions of Sunni Islam

and its caliph against both the infidel and the heretic. These functions were exercised most successfully by

rulers of Turkish origin. Most of the regional dynasts imitated the cAbbasid Caliphs, and buttressed their

own power, by maintaining regiments of Turkish slave guards (Arabic sing. ghulam, mamluk; Persian

banda) from the pagan steppelands of Central Asia. Ghulam status, it must be emphasized, bore none of the

degrading connotations associated with other kinds of slavery. The Turkish peoples enjoyed a particularly high reputation for martial skill and religious orthodoxy, and ghulams were highly prized by their masters,

receiving both instruction in the Islamic faith and a rigorous military training.2 Nor was such confidence

misplaced: as we shall see, the forging and preservation of an independent Muslim power in India were to

be in large measure the work of Turkish slave commanders and their own ghulams.

Mahmud's dynasty, the Ghaznawids or Yaminids (352-582/962-l186), was of Turkish stock; its

effective founder, Mahmud's father Sebuktegin, had been a Turkish slave commander. At its greatest

extent, the Ghaznawid empire embraced an area from Rayy and Isfahan in Persia as far as Hansi in the

eastern Panjab. Mahmud himself, who conducted no less than seventeen expeditions against pagan Indian

rulers and who also rooted out the Isma'ilis from the cities of Multan and Mansuira, was rewarded by the

'Abbasid Caliph for his services to Sunni Islam with the laqab of Yamiin al-Dawla ('Right Hand of the

State').3

Turks did not enter the civilized lands of Islam only through the slave traffic, however. They also

came in as free men, in the large-scale migrations or invasions of recently converted nomadic tribal groups

from Central Asia; and one such clan, the Seljuks, who originated among the Ghuzz (Oghuz) confederacy

north of the Aral Sea, created in the second half of the eleventh

2 C. E. Bosworth, 'Barbarian incursions: the coming of the Turks into the Islamic world', in D. S. Richards

(ed.), Islamic civilisation 950-1150 (Oxford, 1973), 1-16 (especially 4-10), and repr. in Bosworth, The

medieval history of Iran, Afghanistan and Central Asia (London, 1977). D. Ayalon, 'The European-Asiatic

steppe: a major reservoir of power for the Islamic world', in Trudy XXV mezhdunarodnogo kongressy

vostokovedov, Moskva 1960 (Moscow, 1963, 5 vols.), II, 47-52; idem, 'Preliminary remarks onthe Mamluk military institution in Islam', in V. J. Parry and M. E. Yapp (eds.), War, technology and society in the

Middle East (Oxford and London, 1975), 44-58; both repr. in Ayalon, The Mamluk military society

(London, 1979).

C. E. Bosworth, 'The imperial policy of the early Ghaznavids', IS 1 (1962), part 3, 49-82, repr. in his

Medieval history. For a brief survey of the dynasty, see B. Spuler, 'Ghaznawids', Enc.Isl2. The standard

works are Bosworth, The Ghaznavids. Their empire in Afghanistan and eastern Iran 994:1040, 2nd edn

(Beirut, 1973); idem, The later Ghaznavids, splendour and decay: the dynasty in Afghanistan and northern

India 1040-1186 (Edinburgh, 1977).

~century an empire that comprised the whole of Persia, Iraq and Syria. In 344/ 1055 the Seljuk leader

entered Baghdad and took the caliph under his protection, receiving in return the new and exalted style of

Sultan. The Seljuks had already defeated the Ghaznawid amir, Mahmud's son Mas'ud I (431/ 1040); and his

successors, who assumed the title of sultan as a counterblast to Seljuk pretensions, were gradually driven

from their lands in eastern Persia. In 511/1117 the Ghaznawid Bahram Shah was enthroned with the

assistance of the great Seljuk Sultan Sanjar, who dominated the eastern Iranian world from his base in

Khurasan. The Ghaznawids thereby became tributary to the Seljuks; even so, it was not the Seljuks who

would destroy them.

Mu'zz al-Din's family, the Shansabanids, originated among the petty princes (muluk; sing, malik)

of Ghur, the mountainous region east of Herat.4 Reduced to tributary status first by Mahmud of Ghazna and

later by the Seljuks, they found their opportunity at a time of renewed upheavals in the Iranian world. In the 1120s, by one of the same processes in the eastern Asiatic steppe that would bring conquering Mongol

armies westwards in the thirteenth century, the Qara-Khitan (or -Khitai), a semi-nomadic people of

probably Mongolian stock and under the leadership of a Buddhist ruling dynasty, moved into Turkestan

and Transoxiana (Ma wara' al-Nahr) 'and established their hegemony over the Muslim rulers there. Sanjar

was defeated in 536/1141, and in the middle of the century, under pressure from fresh waves of Ghuzz

tribesmen dislodged from their homelands by the Qara-Khitan, his empire collapsed. The Ghuzz also

wrested Ghazna from Bahram Shah's son and successor, Khusraw Shah, and obliged him to fall back on

Lahore (Lahawr), the administrative centre of his Indian territories. The Shansabanids; who had for some

years been embroiled in a feud with the Ghaznawids, were the ultimate beneficiaries of these

developments. Already, in c. 544/1150, the Shansabanid prince cAla" al-Din Husayn had temporarily

expelled Bahram Shah from Ghazna and sacked the city, thereby winning the undying sobriquet of Jahansuz ('World-Burner'); and he took for himself the title of sultan and the ceremonial parasol (chatr)

affected by the Seljuk sovereigns. It was Husayn's nephew, Ghiyath al-Din Muhammad b. Sam (558-

599/1163-1203), who expelled the Ghuzz from Ghazna in 569/1173-4 and installed there his younger

brother Mucizz al-Din (formerly Shihab al-Din) Muhammad.

Under Ghiyath al-Din and Mucizz al-Din, who throughout cooperated more or less harmoniously,

the Shansabanids - or Ghurids, as we may now call them, since they had reduced to subordinate status the

other maliks of

4A. Maricq and G. Wiet, Le minaret de Djam. La decouverte de la capitale des sultans Ghorides (XIF-XIII"

siecles) (Paris, 1959): 31-44 contain a historical survey of the dynasty down to c. 1200; more generally, see

C. E. Bosworth, 'Ghurids', Enc.Isl2; A. D. H. Bivar, 'Ghur', ibid. For what follows, see also Bosworth, Later Ghaznavids, 111-22; idem, 'The political and • dynastic history of the Iranian world (A.D. 1000-1217)', in

J. A. Boyle (ed.), The Cambridge history of Iran, V. The Saljuq and Mongol periods (Cambridge, 1968),

157-66, 185-92.

~the region - emerged as one of the great powers of the eastern Islamic world. Their principal seat was the

fortress of Firuzkuh, identified by Andre Maricq in 1957 with ruins at Jam on the middle Hari Rud, some

200 km. east of Herat; Ghiyath al-Din's authority was recognized by branches of the dynasty which ruled at

Bamiyan, Madin and Jurwas. His chief rivals were the rulers of Khwarazm on the lower Oxus (Amu-

darya), who belonged to a dynasty founded by a Turkish ghulam and who like the Ghurids were erstwhile

subordinates of the Seljuk Sultan. But the Khwarazmshahs suffered from two disadvantages that did not

afflict the Ghurids. One was the overlordship of the heathen Qara-Khitan to their rear (although their military support could on occasions prove welcome); the other was the hostility of the cAbbasid Caliph al-

Nasir li-Dini'llah (575-622/1180-1225). Encouraged by the caliph, from whom he obtained the title Qasim

Amir al-Mu'minin ('Partner of the Commander of the Faithful'), Ghiyath al-Din engaged in a duel for

Khurasan with the Khwarazmshahs, in which, prior to his death in 599/1203, the Ghurids definitely had the

better of it. Mucizz al-Din, who like Ghiyath al-Din bore the title of sultan,'ably seconded his brother's

efforts; but he also looked eastwards.

Early Muslim India

For the first few centuries after Muhammad b. Qasim's conquest of Sind, the frontier in India between the Islamic world - the Dar al-Islam ('Abode of Islam') - and pagan territory - the war-zone or Dar al-Harb -

had remained relatively static. The early Muslim governors of Sind engaged in' holy war (jihad) against

their Hindu neighbours, despatching periodic expeditions as far afield as Kashmir or Malwa.5 But until the

first decades of the tenth century, Muslim expansion eastwards was effectively barred by the powerful

Gurjara-Pratihara dynasty, which dominated northern India from its capital at Kanauj (Qinnawj) on the

Ganges. Mahmud of Ghazna undoubtedly benefited from the eclipse of this empire and the division of its

territories among a number of warring successor-states. Many of his victories in India achieved nothing

more than the acquisition of unheard-of quantities of plunder: Hindu cities were sacked, notably the great

seaport of Somnath in Gujarat (416/1025-6), their temples looted and golden idols piously smashed to

pieces and carried off to Ghazna to replenish Mahmud's treasury. But for all their swashbuckling character,

one result of the Ghaznawid amir's activities was the acquisition for Islam of a new foothold in the western

Panjab.6 5 For a convenient list of campaigns, see J. F. Richards, 'The Islamic frontier in the east: expansion into

South Asia', South Asia 4 (1974), 94-8; and on early Muslim India more generally, Andre Wink, Al-Hind:

the making of the Indo-Islamic world, I. Early medieval India and the expansion of Islam, 7th-llth centuries

(Leiden, 1990), esp. chap. 4.

6 M. Nazim, The life and times of Sultan Mahmud of Ghazna (Cambridge, 1931), chapter 8.

~

Following their expulsion from eastern Persia, the Ghaznawids were increasingly confined to their lands in present-day Afghanistan, Makran and Sind and to their conquests in India. Within the subcontinent they

forfeited some of Mahmud's gains. Hansi, for example, was wrested from them by a coalition of Hindu

princes in 435/1043; and Multan again passed into the hands of the Isma'ilis.7 But the dynasty was by no

means moribund. The reigns of Ibrahim (451-492/1059-99) and of his son Mas'ud III (492-508/1099-1115)

were characterized by the continuing prosecution of the traditional mission in India. It is in 1090 that we

first encounter, in an inscription of the Gahadavala king of Kanauj, the mysterious Turushka-danda, a tax

designed either to finance the struggle against the Muslims or to meet their demands for tribute. According

to the chronicler Juzjani, Mas'ud Ill's military chamberlain (hajib) Toghategin mounted a raid which

penetrated beyond the Ganges and further east than any Muslim incursion since the time of Mahmud. The

dynasty did not abandon military exploits even in an era of decline. Bahram Shah is said to have conducted

holy wars (ghazuha) in India, and his grandson Khusraw Malik appears to have fought against Hindu

powers not long before the truncated Ghaznawid Sultanate was finally overwhelmed by the Ghurids.8

The Ghurid conquests

We possess a number of sources for the Ghurid campaigns of conquest and for the emergence of an

autonomous Muslim power in northern India. The Tabaqat-i Nasirl of Minhaj al-Din b. Siraj al-Din

Juzjani, completed in Delhi in 658/1260, is a general history of the Islamic world in twenty-three sections

(tabaqat), of which sections 19 and 20 deal with the Ghurids and their immediate successors in India. A

precious source for the mid-thirteenth-century Delhi Sultanate, it is of less value for events in India prior to

623/1226 when the author was still resident in Ghur.9 Of the earlier works composed in India, Hasan-i

Nizami's florid and verbose Taj al-Ma'athir, begun in 602/1205-6 but completed after 626/1229, is the

nearest thing we have to a narrative of events. This work, which opens with Mucizz al-Din's great victory at Tara'in in 588/1192, may have drawn upon the victory despatches (fath-namas) of Mu'izz al-DIn's slave

general Aybeg. For

7 J. Burton-Page, 'Hansi', Enc.Isl.2; Bosworth, Later Ghaznavids, 32-3. AH, 252-4.

8 Toghategin: TN, I, 240 (tr. 107). Bahram Shah: ibid., I, 241 (tr. 110). Khusraw Malik: AH, 272; partial tr.

I.M. Shafi, 'Fresh light on the Ghaznavids', IC 12 (1938), 218. Turushka-danda: Bosworth, Later

Ghaznavids, 67; but for a discussion of the possible meanings, Lallanji Gopal, The economic life of

northern India, c. A.D. 700-1200, 2nd edn (Delhi, 1989), 48-52. See also Bosworth's comments, Later

Ghaznavids, 61-7, 84-6, 125-6, on Ghaznawid vigour, together with the evidence accumulated in A. B. M. Habibullah, The foundation of Muslim

9 rule in India, 2nd edn (Allahabad, 1961), 57-60.

On the author, see K. A. Nizami, On history and historians of medieval India (New Delhi, 1983), 71-93.

~

all its defects, it can claim to be the first chronicle written in the Delhi Sultanate.10 A fairly skeletal outline

from 588/1192 down to the events of 602/1206, following Mu'izz al-Din's murder, is to be gleaned from the

prologue to Fakhr-i Mudabbir's Shajara (or Bahr) al-Ansab, composed at Lahore shortly afterwards;

although it does supply dates for certain events that are not given elsewhere. Regrettably, Fakhr-i Mudabbir's later work, Adab al-Harb wa'l-Shaja'a, a military and administrative treatise presented to the

first Delhi Sultan, Iltutmish, in c. 630/1232, does not include among its numerous anecdotes any pertaining

to more recent decades.11 Similarly, only a small proportion of the material relating to India in the Jawami'

al-Hikayat, a large collection of historical anecdotes compiled by cAwfi in Delhi (c. 628/1230-1), dates

from the post-Ghaznawid era.12 It is fortunate that events on this distant frontier made a powerful

impression in Islam's heartlands. We should be much less well informed about the Ghurid campaigns were

it not for the al-Kamilfi'l-Ta'rikh, a general history by Ibn al-Athir (d. 630/1232), who wrote in the Iraqi

city of al-Mawsil (Mosul); though where he obtained most of his information was as great a mystery to at

least one contemporary as it fs to us.13

Once installed at Ghazna, Mu'izz al-Din was not slow to appropriate the Ghaznawids' role as the standard-bearer of orthodox Islam in the subcontinent. As Mahmiid had done, he made war on the Isma'ilis, who had

reestablished themselves in Multan, and captured the city (571/1175-6); the evidence suggests that although

the Sumra princes at Daybul in the Indus delta, whom he attacked in 578/1182-3, were of Indian stock, they

too may have been Ismacili sympathizers. Certainly he is praised for his warfare against the Shfls.14 But the

annexation of the remaining Ghaznawid territories was undoubtedly his principal goal. A series of

campaigns from 577/1181-2 onwards secured first tribute from Khusraw Malik and then, in

10 A critical edition is very much to be desired. Unless otherwise stated, references are to 1OL Persian ms.

15 (Ethe, no. 210). The standard version ends in 614/1217, although in the last century Sir Henry Elliot

utilized a copy (since lost) that went down to 626/1229: abstract translated in ED, II, 240-2. For a useful

summary of the main recension, see S. H. Askari, 'Taj-ul-Maasir of Hasan Nizami', PUJ 18 (1963), no. 3,

49-127; on the author, Nizami, On history and historians, 55-70. 11 M. S. Khan, 'The life and works of Fakhr-i Mudabbir', IC 51 (1977), 127-40. E. Denison Ross, 'The

genealogies of Fakhr-ud-din Mubarak Shah', in T. W. Arnold and R. A. Nicholson (eds.), c Ajab-Nama: a

volume of oriental studies presented to Edward G. Browne (Cambridge, 1922), 392-413.

12 On the author's life, see M. Nizamu'd-din, Introduction to the Jawdmfu'l-hikayat, GMS, ns, VIII

(London, 1929), 3-20.

13 D.S. Richards, 'Ibn al-Athir and the later parts of the Kamil: a study of aims and methods', in D.O.

Morgan (ed.), Medieval historical writing in the Christian and Islamic worlds (London, 1982), 84-5.

14 SA, 19-20. Habibullah, Foundation, 36-7. S.H. Hodivala, Studies in Indo-Muslim history (Bombay, 1939-

57, 2 vols.), I, 141. For Daybul, see S. Qudratullah Fatimi, 'The twin ports of Daybul', in Hamida Khuhro

(ed.), Sind through the centuries (Oxford and Karachi, 1981). 97-105; Wink , Al-Hind, I, 181-3.

~

582/1186, the capitulation of Lahore. Khusraw Malik was sent to Ghiyath al-Din and later put to death in

captivity.

Confronting the Ghurid ruler now were a number of major Hindu powers, for which the designation 'Rajput' (not encountered in the Muslim sources before the sixteenth century) is a well-

established anachronism.15 Chief among them was the Chahamana (Chawhan) kingdom of Sakambhari

(Sambhar), which dominated present-day Rajasthan from its capital at Ajmer; it included much of the

territory between the Sutlej and the Yamuna, and under Prthvlraja III (the 'Rai Pithura' of Muslim writers)

claimed paramountcy throughout India north of the Vindhya mountains. Junior branches of the dynasty

ruled at Nadol and at Jalor, and Delhi (Dilli, Dihli), under its Tomara princes, had been tributary to the

Chawhans since the middle of the twelfth century. Chawhan supremacy was of relatively recent date,

however, having been won in the teeth of strenuous opposition from the Chaulukyas, who reigned over

Gujarat from their capitaliat Anhilwara (Nahrwala; now Patan) and still nurtured designs on southern

Rajasthan. To the east, the Chawhan state bordered on the 'Gahadavala kingdom of Kanauj (Qinnawj; the

ancient Kanyakubja), which dominated much of the modern province of Uttar Pradesh, and the Chandella

kingdom of Jejakabhukti (modern Bundelkhand), centred on Kalinjar. In the 1180s the Chandellas were under pressure from both the Gahadavalas and the Chawhans, and forfeited some of their western territories

to Prthvlraja III. The Gahadavala kingdom, on the other hand, was also busily expanding into Bihar, where

it contested the debris of the defunct Pala empire with the Sena dynasty of western Bengal.16 In all these

states, there existed a quasi-feudal hierarchy in which the kings (rajas, called rais by the Muslim invaders)

received military service, in return for grants of land, from subordinate chieftains, called ranakas (or

sometimes thak-kuras), who in turn conferred estates on their own cavalry commanders, the rautas (from

Skr. rajaputras) or nayakas; these two lower levels are the ranas and rawats respectively of the Muslim

sources.17

15 For this term, see B. D. Chattopadhyaya, 'Origin of the Rajputs: the political, economic and social

processes in early medieval Rajasthan', IHR 3 (1976), 59-82, repr. in his The making of early medieval India (Oxford and Delhi, 1994), 57-88.

16 See generally H. C. Ray, The dynastic history of northern India (Calcutta, 1931-5, 2 vols.), chaps. 6

(Senas), 8 (Gahadavalas), 11 (Chandellas), 15 (Chaulukyas) and 16 (Chahamanas); also Dasharatha

Sharma, Early Chauhan dynasties, 2nd edn (Delhi, 1975); R. C. Majumdar, Chaulukyas of Gujarat

(Bombay, 1956); Roma Niyogi, The history of the Gahadavala dynasty (Calcutta, 1959); A. Banerji,

'Eastern expansion of the Gahadavala kingdom', JAS, 4th series, 5 (1963), 105-11; N. S. Bose, History of

the Chandellas (Calcutta, 1956); and R. K.

1? Dikshit, The Candellas of Jejakabhukti (New Delhi, 1977).

R- S. Sharma, Indian feudalism: c. 300-1200 (Calcutta, 1965), especially chap. 5. Pushpa Prasad (ed.),

Sanskrit inscriptions of Delhi Sultanate 1191-1526 (Delhi, 1992), 56-7 (no. II:5), 58-71 (no. 11:6), 78-9 (no. 11:9), 80-9 (no. II:11). For examples from Muslim sources, see SA, 33 (with RATGAN in error for

RANGAN); Taj, fols. 137a, 150a; and inter alia the 'celebrated rawats of TN, II, 65 (tr. 828).

~

Significant gains at the expense of these Hindu powers were deferred until after Mucizz al-Din's

annexation of the Ghaznawid territories, which brought him control of the more northerly routes via

Peshawar (Parshawar) and the Khyber Pass. Indeed, his earliest incursion into the Dar al-Harb had ended in

disaster. An attack in 574/1178-9 on the Chaulukya kingdom by way of lower Sind resulted in a heavy

defeat for the Ghurid Sultan near Mount Abu. Subsequently, at a date which is variously given as

583/1187-8 or 587/1191, he invaded the eastern Panjab and established a garrison at Tabarhindh. But he was routed at Tara'in by a large Hindu force under Prthvlraja and his subordinate, Govindaraja of Delhi,

and obliged to retire to Ghazna; Tabarhindh was recovered by the Hindus.18 When Mu'izz al-Din returned

in 588/1192, however, and again offered battle near Tara'in, he won a crushing victory, in which Prthvlraja

was captured and Govindaraja killed. The victory at Tara'in seems to have constituted a turning - point

intwo respects. Firstly, the Hindu chiefs of the eastern Panjab undertook topay tribute to Mucizz al-Din.19

And in the second place, it is from this moment that we can date the establishment of a permanent Muslim

force in the region, at Indraprastha (Indrapat), near Delhi.20 But direct Muslim rule was not imposed on a

uniform basis. While the great Chawhan fortress of Ranthanbor was occupied, Ajmer was left in the

possession of Prthvlraja, now Mu'izz al-Din's client; and following his execution for some act of duplicity

shortly afterwards, it was conferred on his son. Similarly, Delhi was granted to Govindaraja's successor as a tributary prince.21 This pattern was to be followed many times in other regions conquered by the Muslims.

Mu’izz al-Din continued to move down from Ghazna into India for each cold season and to take

charge of the war against the infidel. In 590/1194 it was the turn of the Gahadavalas, whose king

Jayachandra (the 'Jaychand' of Muslim authors) was defeated and slain by Mucizz al-Din in the.vicinity of

Chandawar (Chandawal, near Etawa); the Ghurid army looted his treasury at Asi (Asm) and occupied

Banaras (now Varanasi). In 592/1196 the sultan headed an expedition which secured the fortress of Thangir

(Tahangarh, fifteen miles south of the later city of Bhayana) from the

18 Habibullah, Foundation, 60-1. TN, I, 398-400 (tr. 457-64, 466), where this engagement is dated in the

year preceding the second battle of Tara'in. IA, XI, 113-14/172-3, 371-2/ 561-2, describes the campaign

twice (cf. XI, 115/174): in the second account, he dates the episode in the latter half of 583 (ended 1 March 1188), and this is confirmed at XII, 59/91. 19 Ibid., XI, 115/174, wa-iltazamu lahu bi'l-amwal. Taj, fol. 50b, for the chieftains of the Delhi region

specifically. 20 Ibid., fol. 51a. 21 Habibullah, Foundation, 61-2. On the coinage believed at one time to reflect Prthviraja's client status, see

now P. N. Singh, 'The so-called joint issue of Muhammad bin Sam and Prithviraja III: a reappraisal', JNSI

50 (1988), 120-3; John S. Deyell, Living without silver: the monetary history of early medieval North India

(Oxford and Delhi, 1990), 267-9. That Delhi was thus subjected in two stages may help to explain the

conflicting dates given for its capture in the sources, on which see Muhammad Aziz Ahmad, Political

history and institutions of the early Turkish empire of Delhi (1206-1290 A.D.) (Lahore, 1949), 129 n.l.

~

Chandellas, and allowed the rai of Gwaliyor to buy him off with tribute. But otherwise Mu'izz al-Din

appears to have played a relatively limited role in the extension of Muslim power. After the death of his

brother Ghiyath al_Din (599/1203), his energies were largely absorbed by developments in Khurasan,

where the Khwarazmshah 'Ala' al-Din Muhammad b. Tekish sought to recover territories previously lost to

the Ghurids. In 601/1204 Mu'izz al-Din invaded Khwarazm itself, only to suffer a decisive defeat by the

shah's Qara-Khitan overlords at Andkhud (now Andkhoi).22 In these circumstances, the Ghurid Sultan

seems to have relied in India increasingly on his Turkish slave lieutenants.

The Ghuris were a people of the hills. Traditionally they fought on foot, and Juzjani has left us a

description of their characteristic method of warfare, which involved the use by each soldier of a protective

screen called a karwa, made of raw bullock-hide and filled with a dense wadding of cotton.23 It is true that we also encounter mounted Ghuri warriors, like the 1200 horsemen from Tulak who briefly garrisoned

Tabarhindh following Mu'izz al-Din's first invasion of the eastern Panjab;24 but they were probably in short

supply, and the sultans' expansionist designs required access to larger numbers of cavalry. As the empire

expanded to the west, they supplemented their forces with warriors from various parts of Khurasan:

Khurasanls are found under Mu'izz al-DIn's banner, for instance, in the final thrust against the Ghaznawids

and in his assault on Prthviraja, and later among the troops who entered Lahore with Aybeg in 602/1206.25

In addition, Ghuzz warriors appear in the army of Ghazna in the period following Mu'izz al-Din's death,

and the Ghurid sultans, like their Ghaz-nawid precursors, recruited tribal cavalry from among the Khalaj, a

nomadic people in the garmsir ('hot') regions of Bust and Zamindawar, who may have been of Turkish

stock but would in time become assimilated to the neighbouring Afghans.26 Only late authors mention the

Afghans proper, who were as yet confined to the Sulayman range (consequently known at this time as kuh-i Afghan, 'the Afghan mountains') and who had accompanied Ghaznawid campaigns, as serving at Tara'in.27

22 W. Barthold, Turkestan down to the Mongol invasion, 3rd edn by C. E. Bosworth, GMS, ns, V (London,

1968), 349-51. Bosworth , 'Political and dynastic history', 164-5.

23 TN, I, 343 (tr. 352 -3). The kdrwa is also listed in AH, 423, among the equipment required to

conduct a siege.

24 TN, I, 399 (tr. 458); and for an earlier reference to mounted Ghuris, see I, 355-6 (tr. 372-3).

25 IA,XI, 110/168, 113/172.SA,33.

26 Ghuzz: IA, XII, 144/219. On the ethnicity of the Khalaj, see V. Minorsky, 'The Turkish dialect of the

Khalaj', BSOS 10 (1940), 426-32, repr. in his The Turks, Iran and the Caucasus in the middle ages

(London, 1978); C. E. Bosworth, 'Khaladj, i. History', Enc.Isl.2; C. E. Bosworth and Sir Gerard Clauson,

'Al-Xwarazmi on the peoples of Central Asia', JRAS (1965), 6, 8-9, repr. in Bosworth, Medieval history.

But for a different view, cf. Irfan Habib, 'Formation of the Sultanate ruling class of the thirteenth century',

in Habib (ed.), Medieval India 1.

27 Researches in the history of India 1200-1750 (Oxford and Delhi, 1992), 2-3 and n.12. G. Morgenstierne,

'Afghan', Enc.Isl2. For Afghan warriors under the Ghaznawids, see

~

The Ghurid dynasty grew familiar with the disadvantages of relying exclusively on such forces. The

nomads were proverbially volatile. When cAla' al-Din Husayn 'Jahansuz' did battle with Sanjar in

547/1152, the issue was decided by some 6000 Khalaj, Ghuzz and other Turkish nomads in his army who

went over to the Seljuk Sultan. For Mu'izz al-Din, even the Ghuris did not prove invariably trustworthy.

During his first Tara'in campaign, according to Ibn al-Athlr, his Ghuri troops left him in the lurch, for

which the commanders were severely disciplined; and he continued to harbour resentment against them for

some years.28 Such considerations, as well as the numerous precedents furnished by other Muslim

dynasties, may have encouraged the later Ghurids to amass bodies of Turkish ghulams. Turkish slaves appear at Ghiyath al-Din's court at an early date, and JuzjanI tells us that Mucizz al-Din was especially keen

to acquire them.29 Despite insubordination on the part of one or two ghulam officers in India in the wake of

the sultan's defeat at Andkhud in 601/1204, his confidence was in large measure justified. At Andkhud

Mu'izz al-Din's personal slaves remained with him in the thick of the conflict, and it was one of them who

at length virtually carried him from the field for the sultan's own safety.30 Professor Irfan Habib has shown

how he took care to promote his ghulams (called 'Mu'izzis', from his own laqab) particularly to

administrative and military office in his own territories, Ghazna and India, in contrast with the older Ghurid

lands.31

The principal credit for the Ghurid conquests in the eastern Panjab and beyond is given in the

sources to one ghulam lieutenant, Qutb al-Din Aybeg. It was Aybeg who frustrated Chawhan revanchism

under Prthvira-ja's brother Hariraja ('Hiraj'); who in 589/1193 took possession of Delhi on the pretext of its ruler's treacherous designs; and who in 593/1197 defeated the Chaulukyas at Mount Abu, thereby avenging

his master's humiliation of almost twenty years before. Within the crumbling empire of the Gahada-valas,

Aybeg took the fortresses of Mirat (Meerut) in 588/1192, Kol (near modern Aligarh) in 591/1194, Bada'un

(Budaon) in 594/1198, and Qinnawj (Kanauj) in 595/1199. Gwaliyor surrendered to him in 597/1200-1,

and in 599/1203 he occupied Kalinjar, capital of the Chandella king Pararnardi-deva (Hasan-i Nizami's

'Parmar').32

During these years other elements were carrying Muslim arms even deeper into India. A Khalaj

warrior named Muhammad b. Bakhtiyar had secured a base in Awadh, from where he mounted regular

plundering expeditions into the Hindu tracts of Maner and Bihar. He grew strong

Andre Wink, Al-Hind: the making of the Indo-Islamic world, II, The Slave Kings and the Islamic conquest,

I1 th-13th centuries (Leiden, 1997), 116-18. 28 TN, I, 346 (tr. 359). IA, XI, 114/173,371-2/561-2.

29 Ghiyath al-Din: TN, I, 354-5, turkan-i khass (tr. 371). Mueizz al-Din: ibid., I, 410 (tr. 497).

30 Ibid., I, 403 (tr. 476-8). 31 I. Habib, 'Formation', 4-7. See also Wink, Al-Hind, II, 141. 32 For these

campaigns, see Habibullah, Foundation, 62-9; HN, 156-90.

~

enough first to take the city of Bihar and then to attack the Sena kingdom in western Bengal. In the middle

of Ramadan 601/early May 1205 Nudiya, the capital of king Laksmanasena ('Lakhmaniya'), was captured

and sacked, and the king himself put to flight. Muhammad b. Bakhtiyar was murdered in c. 602/1206

following a disastrous campaign somewhere in Assam (Kamrtup). Although he acknowledged the Ghurid

Sultan as his master and conveyed a proportion of the plunder to Aybeg, he acted independently, without

the benefit of direction - or even, as far as we can tell, reinforcements - from Ghazna.33 These operations,

the fame of which would reach the ears of Ibn al-Athir in Iraq and would cause a later author to give the

Khalaj alone the credit for the Muslim conquests, reduced for Islam a considerable tract in the Ganges

basin where Mucizz al-Din's forces had not penetrated.34

The news of Mucizz al-Din's defeat at Andkhud in 601/1204 provoked a rebellion by one of his lieutenants, who seized Multan, and a more formidable rising by the Hindu Khokhars and the people of the

Salt Range (Kuh-i Jud); and his last years were taken up with their suppression. On his murder in 602/1206

(probably by Isma'ilis from Khurasan), his empire fell apart. He left no son, and his vast inheritance was

disputed by his relatives and slaves and his enemy the Khwarazmshah cAla' al-Din Muhammad, who

repudiated Qara-Khitan overlordship and annexed the Ghurid territories in Khurasan. Ghazna was occupied

by the late sultan's senior ghulam, Taj al-Din Yildiz; but in the years 611-12/1214-16 he and the various

Ghurid princes were alike overwhelmed by the Khwarazmshah, not long before the Khwarazmian empire

was destroyed (618-20/1221-3) by the advancing Mongols of Chinggis Khan.35 The Indian provinces

meanwhile went their own way. On learning of his master's death in 602/1206, Aybeg advanced from Delhi

and took up residence at Lahore, where he established himself as ruler. When Aybeg died in a polo

(chawgan) accident in 607/1210-11, his ghulam Iltutmish was invited into Delhi from Bada'un by a party in the city, and set himself up as ruler in opposition to Aybeg's her.36 Aybeg's action marks the emergence of

an independent Muslim power in India; that of Iltutmish, the creation of the Delhi Sultanate, which will be

the subject of the next chapter.

23 Taj,fol. 186a-b. T , I, 423-7 (tr. 551- 4, 560). For the Nudiya campaign, see Habibullah, Foundation, 69-

74; and for the date, Parmeshwari Lal Gupta, 'On the date of the Horseman type coin of Muhammad b.

Sam', JNSI38 (1976), no. 1, 81-7.

34 IA, XI, 115/174. For the role of the 'Qalaj' in the conquest of 'Hindustan', see Ibn Sacid al-Maghribi (d.

673/1274 or 685/1286), Kitab al-Jughrafiyya, ed. Ismacil al-cArabi (Beirut, 1970), 163: on this author, see

Gilles Potiron, 'Un polygraphe andalou du XIIIe siecle', Arabica 13 (1966), 142-67.

35 P. Jackson, 'The fall of the Ghurid dynasty', in Carole Hillenbrand (ed.), Festschrift for Professor

Edmund Bosworth (Edinburgh; forthcoming). 36 Habibullah, Foundation,88-92.

~Reasons for the Ghurid victories

It is easier to chronicle the triumphs of the Ghurid armies in India than to account for them; and certainly

no satisfactory explanation is forthcoming in the sources. For the four Muslim writers who notice these

events, it is enough that God grants victory to the sultan and his forces. Any analysis of the causes of

Muslim success, therefore, rests on fragmentary evidence, and our conclusions can only be speculative.

We must first discuss one hypothesis which has at times been adduced in explanation of the Muslim conquest of northern India at the turn of the twelfth century. Drawing on the observations about the

caste system to be found in the work of the eleventh-century Muslim writer al-Biruni, the late Professor

Mohammad Habib suggested that the resistance of Hindu rulers, when confronted by the invading Ghurid

armies, was undermined in two respects. First, the caste system seriously impaired the military

effectiveness of the Hindu kingdoms. It restricted participation in war to the warrior caste, the kshatriyas,

and the principle of untouchability required them, on the eve of battle, to perform numerous tasks that

would otherwise naturally have fallen to those of menial rank. The second disadvantage allegedly imposed

on the Hindu states by the caste system was its effect upon the cohesiveness of the subject population.

Islam preaches equality. Faced with this liberating message (the argument runs), the urban masses could

not but draw the contrast with the social shackles that bound them and throw in their lot with the newcomers. Habib thus concluded, in words that have attained a certain notoriety, that 'this was not a

conquest so-called. This was a turnover of public opinion, a sudden one no doubt, but one which was long

overdue.'37

Although these ideas are appealing at first sight, they do not withstand closer scrutiny. As far as

military effectiveness is concerned, it has been pointed out both that Hindu armies included members of

other castes, such as vaisyas and sudras, and that al-Biruni's Brahman informants may have exaggerated the

effectiveness of the caste regulations.38 Regarding the question of liberation, we need to know far more

than we do about the perceptions that the lower-caste Hindu populace had of their situation and the

message (if any) preached by the invading Muslim troops. At the risk of stating the obvious, it might be

pointed out that a recognition of one's low social status, particularly when sanctioned by religious laws, and

an urge to 37 M. Habib, 'Introduction to Elliot and Dowson's History of India, vol. II', in K. A. Nizami (ed.), Politics

and society during the early medieval period. Collected works of Professor Mohammad Habib (New Delhi,

1974, 2 vols.), I, 59-74 (72 for the quotation). See also Aziz Ahmad, Studies in Islamic culture in the Indian

environment (Oxford, 1964), 82.

38 Other warriors: Prabha Dixit, 'Prof. Mohammad Habib's historical fallacies', in Devahuti (ed.), Bias in

Indian historiography (Delhi, 1980), 205. Caste regulations: S. Digby, review of Habib's collected works,

in BSOAS 39 (1976), 457.

~improve it do not necessarily - in a society untouched by the eighteenth-century European Enlightenment - go hand in hand.39 Nor can the liberation that the Muslim conquerors offered to those who sought to

escape from the caste system be taken for granted. The evidence for widespread conversion to Islam at the

turn of the twelfth century simply does not exist. That such deliverance was in fact on offer seems

improbable in view of our knowledge of the early centuries of Muslim rule in Sind, which is somewhat

fuller than it is for conditions in the newly conquered Indian territories of the Ghurids.

.The principal source for the Arab reduction of Sind in the early eighth century is the Chach-

Nama, a Persian work composed in c. 613/1216-17 but purporting to be a translation of an earlier, Arabic

history. It alleges that Muhammad b. Qasim, the conqueror of Sind, learned of the disabilities imposed on a

local people, the Jats, in the era of the deposed Brahman dynasty. One was that the Jats were to take dogs

with them whenever they went out of doors, in order that they might be recognized. Muhammad b. Qasim

ordered that such disabilities continue in force. That they did so emerges from a passage in the Futuh al-Buldan of al-Baladhuri (d. 279/892), in which a caliphal governor of Sind in the late 830s is said to have

required the Jats, when walking out of doors in future, to be accompanied by a dog. The fact that the dog is

an unclean animal to both Hindu and Muslim made it easy for the Muslim conquerors to retain the status

quo regarding a low-caste tribe. In other words, the new regime in the eighth and ninth centuries did not

abrogate discriminatory regulations dating from the period of Hindu sovereignty; rather, it maintained

them.40 We have no grounds for supposing that the response of the late twelfth-century conquerors to the

caste system was any different.

To turn now to other possible explanations for the Ghurid victories, military technology is one

sphere in which the Muslims may have enjoyed some limited superiority. Mucizz al-Din is described in one

Hindu source as 'lord of the north-west, where horses abound',41 and it is accordingly possible that he was able to field a larger cavalry force than his opponents. This question has been examined by Simon Digby

for the era of the Delhi Sultanate proper,42 and will be considered further in subsequent chapters. For the

moment, two other circumstances should be pointed out. One is that Ghur had long been renowned for its

metal deposits and its manufac-

39 A point well made by Friedmann, 'A contribution', 320-1.

40 CN , 33. Baladhuri, Futuh al-Buldan, ed. M. J. De Goeje, Liber expugnationis regionum (Leiden, 1866),

445/ed. S. al-Munajjid (Beirut, n.d.), 544. Friedmann, 'A contribution', 331-2. See alsp the brief remarks in

Irfan Habib, 'Economic history of the Delhi Sultanate -an essay in interpretation', IHR 4 (1977), 297: There is no evidence of any direct assault

41 from the state or the Muslims upon the caste system; nor even of any revolt from within ...'

42 Har Bilas Sarda, 'The Prithviraja Vijaya', JRAS (1913), 279.

s- Digby, War-horse and elephant in the Delhi Sultanate: a problem of military supplies (Oxford and

Karachi, 1971).

~ ture of weapons and coats of mail, commodities that had at one time formed part of the tribute rendered

successively to the Ghaznawids and the Seljuks.43 It is conceivable, therefore, that Mu'izz al-Din drew on a

more plentiful supply of armaments for his Indian campaigns than recent Ghaznawid Sultans (or, for that matter, the Delhi Sultans in the next century). The other important consideration is that Hasan-i Nizami, in

describing the campaigns of Mu'izz al-Din and Aybeg, refers with remarkable frequency' to the Muslims'

use of the crossbow (nawak) and makes great play of the armour-piercing properties of the crossbow bolt.44

It is by no means clear that the Ghurids' Hindu adversaries made such use of the crossbow.45 This

formidable weapon, which was at this very time giving Latin Christian armies a'decisive advantage over

their enemies in the Celtic and Slavic worlds,46 may well have performed a parallel function for the Muslim

invaders of India. But this would hardly explain the victory at Tara'in, gained in the very locality where

success had eluded Mucizz al-Din not long previously.

The particular tactics that the sultan adopted in the second battle of Tara'in may have played a

significant role in his victory. An anecdote in 'Awfi's Jawami" al-Hikayat suggests that Mucizz al-Din exploited the proximity of the enemy's elephants to the horses, whose fear of elephants renders it difficult

to coordinate bodies of both animals in the field. While campfires were lit to dupe Prthvlraja's men into

believing that the entire Ghurid army had bivouacked for the night, the sultan took a division of his troops

round to attack the Chawhan rear. At daybreak he fell upon Prthvlraja's baggage. The rear was pushed

against the elephants, which got out of control, so that the Chawhan army fell into confusion and Prthviraja

was unable even to conduct an orderly retreat.47 On the other hand, in the short account of the battle found

in Juzjani's Tabaqat-i Nasiri and obtained from an eye-witness, Mu'izz al-Din divided his forces.48 While

the centre,

43 TN, I, 335, 346 (tr. 336, 358). See also the tenth-century Hudud al-cAlam, tr. V. Minorsky, GMS, ns, XI,

2nd edn (London, 1970), 110. Athar AH, 'Military technology of the Delhi Sultanate (13-14th C.)', inPIHC

50 (Gorakhpur 1989) (Delhi, 1990), 167. 44 See especially Taj, fols. 81a, 146b, 201a, 229a; also AH, 400, 423 and passim . On the possible role of

the nawak, see Irfan Habib, 'Changes in technology in medieval India', Studies in History 2 (Aligarh, 1980),

26-7; and for a fourteenth-century dictionary definition of nawak, see Muhammad b. Hindu Shah

Nakhchiwani, Sahah al-Furs, ed. cAbd al-‘Ali Ta’ati (Tehran, 1341 Sh./1962), 188. The term originally

denoted the tubular attachment but was later extended to the weapon as a whole: Kalervo Huuri, Zur

Geschichte des mittelalterlichen Geschutzwesens aus orientalischen Quellen (Helsinki and Leipzig, 1941),

105; also the review by Cl. Cahen, in JA 236 (1946), 169; and Cahen, 'Un traite d'armurerie compose pour

Saladin', BEO 12 (1947-8), 153-4.

45 The occasional reference shows that they did possess the nawak: e.g. Taj, fols. 40b, 130b; AH, 247. 46 Robert Bartlett, The making of Europe: conquest, colonization and cultural change 950-1350

(Harmondsworth, 1993), 63-4, 73-4.

47 JH, BN ms. Ancien fonds persan 75, fol. 185 (abstract in ED, II, 200).

48 For what follows, see 7N, I, 400 (tr. 467-8).

~ the baggage and the elephants were kept several miles in the rear, bodies of picked light-armed cavalry

(sawar-i barahna wa-jarlda), totalling 10,000 men, were ordered to harass the enemy in every direction. These are clearly shown a few lines later to have been mounted archers; and the sultan's instructions to

them - to fire from all sides, and then to retreat and maintain a distance between themselves and the enemy

when the Hindu army attempted to charge - are strikingly reminiscent of the tactics of nomadic Turkish

horse-archers such as the Seljiiks when confronted, for instance, by crusading armies in Anatolia and

Syria.49 Professor Nizami was thereby led to assume that these were the tactics which were instrumental in

winning for Islam the north Gangetic plain.50

Yet the fact that the victory was won in part by the techniques in which Turkish nomads excelled should

not blind us to the rest of the evidence. Mu'izz al-Din's armies did not consist overwhelmingly of Turkish

nomads. The force of ten thousand light-armed horsemen was but a fraction of a much greater army

comprising, says Juzjani, 120,000 cavalry with horses wearing armour (bar-gustuwan).51 Even if this figure

is exaggerated, it seems plain that the Ghurid forces at Tara'in were in large measure made up of heavy cavalry. It is these warriors - and not light-armed horse-archers -who are immortalized on the early Muslim

coinage of Bengal as the very symbol of Muslim domination.52 The Moroccan traveller Ibn Battuta, who

reached the Delhi Sultanate in 734/1333, comments on the fact that heavily armoured cavalrymen still

made up the Delhi Sultan's forces.53 It is worth noticing at this juncture that Turkish slave soldiers were

employed as heavy cavalry - that their value to their employers, in other words, did not lie in any attempt to

replicate the tactics traditionally associated with the steppe.54 Such heavily armoured troops would hardly

have mounted the kind of attacks from which crusading armies suffered. Indeed, their performance would

have been more akin to the tactics of the crusaders themselves: a heavy cavalry charge, whose shock effects

on a relatively immobile opponent were renowned throughout the Near East.55 If 'Awfi's

49 Walter E. Kaegi, Jr, 'The contribution of archery to the Turkish conquest of Anatolia', Speculum 39 (1964), 96-108. R. C. Smail, Crusading warfare 1097-1193 (Cambridge, 1956; 2ndedn, 1995), 75-83.

K. A. Nizami, Some aspects of religion and politics in India in the thirteenth century (Aligarh, 1961), 82;

and in HN, 186.

51 TN, I, 400 (tr. 465-6).

52 CMSD, 6 (no. 3A), 15 (nos. 49F, 49G), 16 (nos. 49H-J), and illustrations at Pis. XXII-XXIV. See also

Richard M. Eaton, The rise of Islam and the Bengal frontier, 1204-1760 (Berkeley and Los Angeles, 1993),

33-5, and the coins there illustrated. The Hindus employed bar-gustuwan horsemen as well: AH, 272; SA,

27.

£ IB, II, 374 (tr. Gibb, 479).

See the observations of Cl. Cahen, 'Les changements techniques militaires dans le Proche Orient medieval

et leur importance historique', in Parry and Yapp, War, technology and society, 121; also Wink, Al-Hind, II,

89. Smail, Crusading warfare, 112-15. Christopher Marshall, Warfare in the Latin East,

~ story embodies authentic detail from the Tara'in campaign, and does not simply describe a stratagem

sometimes adopted by Muslim commanders against Hindu armies in the past, it may possibly echo Mucizz

al-Din's success in rolling the core of Prthviraja's host into a solid mass - against which the light archers

mentioned by Juzjani would have operated to deadliest effect but which would also have presented the

ideal static target for a heavy cavalry attack.

Although we have scarcely any information on numbers, it is conceivable also that Mu'izz al-Din owed his

victories to an increase in the size of his army. The figure of 120,000 cited by Juzjani for the Ghurid army

at Tara'in is clearly designed to make an impact on the reader, and suggests that the sultan had raised an

unusually large force for the invasion. It may already have been reported in the Near East some decades

before Juzjani wrote, and it was to make a sufficiently powerful impression on the Mughal conqueror

Babur, three centuries later, to be included in his memoirs.56 For the army that attacked the Gahadavalas,

the numbers we have are set somewhat lower, at 50,000 heavily armoured cavalry57 - still a massive force,

if the figure is reliable. Many of these troops were probably volunteers: at an earlier date, Ghaznawid

armies operating on the Indian front had been swollen by thousands of men seeking to serve as holy warriors (ghazis).58 Such immigrants would have comprised both Turks and 'Tajiks', as the non-Turkish

population of the Iranian world and Transoxiana were known. The latter category would have included not

merely bureaucrats and the military, but descendants of the Prophet (sadat, sing, sayyid), holy men

(shaykhs) and scholars ('ulama', those well versed in the Holy Law or Qur'anic sciences), like the two

learned (danishmand) brothers from Farghana, troopers under Muhammad b. Bakhtiyar mentioned by

Juzjani, who met one of them at Lakhnawti in 641/1243. One source of recruitment that was certainly

available to Mucizz al-Din was the Khalaj. We know that they were not necessarily light cavalry: the small

force with which Muhammad b. Bakhtiyar stormed into the city of Bihar consisted of two hundred heavily

armed (bar-gustuwan) horsemen. The bands of Khalaj tribesmen. who had flocked to join him only a few

years after the overthrow of Prthviraja are expressly said to

1192-1291 (Cambridge, 1992), 158-63. The Hindus do not seem to have deployed mounted shock combat troops: Wink, Al-Hind, II, 81.

56 Babur-Nama, 479-80. The figure is found in Ibn al-Dawadari, Kanz al-Durar (c. 730/1329), ed. Said cAshur et al, VII (Cairo, 1391/1972), 134. For the date, 590, given here Ibn al-Dawadari cites Ibn al-Sa°I

(d. 674/1276) and Ibn Wasil, but it is impossible to say which of these authors, if either, transmitted the

figure of 120,000. Of Ibn al-Saci's al-Jamf al-Mukhtasar, only the portion covering the years 595-

606/1198-1209 has survived; the work of Ibn Wasil cited is not Mufarrij al-Kurub and must therefore be

his Ta'nkh Sdliln (c. 636/ 1239), found only in an Istanbul ms. which is inaccessible to me.

57 Taj, fol. 119b. 58 See Bosworth, Ghaznavids, 114; Wink, Al-Hind, II, 91-2 n.57.

~ have come 'from the direction of Hindustan' (i.e. the Doab and Awadh), indicating at least that they were

not newcomers to India.59

Holy war, conquest and the infidel

In the space of little more than a decade, the Ghurid armies in India had made striking progress; the

Muslims now held a string of fortresses from which they more or less dominated the north Gangetic plain.

It is important, on the other hand, to recognize the limits of Muslim success. Victory did not necessarily

entail the displacement of Hindu rulers. As we have seen, prthviraja's son was installed as his father's

successor at Ajmer, and Govindaraja's son ruled briefly at Delhi, both as the sultan's subordinates; and

following the victory over Jayachandra Aybeg is said to have installed 'a rana in every direction'.60 The

Ghurid Sultan's position was that of an over-king presiding over a number of tributary princes, the rais and ranas who came, in Hasan-i Nizami's words, 'to rub the ground of the exalted court of Aybeg'.61

Some Muslim triumphs had been merely temporary in character. Aybeg's sack of Nahrwala in 593/1197,

for instance, though dignified by JuzjanI as 'the conquest of Gujarat', had not led, as far as we can tell, to

any acquisition of territory. The consequences of his raid on Malwa in 596/1200 were doubtless equally

ephemeral.62 In the eastern parts of what is now Uttar Pradesh, the Gahadavala kingdom still held out.63

Moreover, although our sources are reluctant to inform us of Muslim reverses, it is clear that of the

strongholds taken by Aybeg some certainly passed back into the hands of Hindu princes, perhaps after his

death, since they had to be retaken by Iltutmish. Further east, Muhammad b. Bakhtiyar's exploits had

secured only the north-western part of Bengal, where Muslim authority now centred on the town of Gawr,

renamed Lakhnawti: eastern Bengal, the region called 'Bang' by the Muslims, remained in the hands of the Sena dynasty.64

Even within the areas over which the Muslims ruled more or less directly, the intensity of their control is

open to question, and it is necessary to

59 TN, I, 423 (tr. 551-2). For this restricted meaning of 'Hindustan', see below, p. 86.

60 Taj, fol. 137a. 61 Ibid., fol. 150a.

62 Gujarat: ibid., fol. 173a, and TN, I, 417 (tr. 516). Malwa: SA, 24, and TN, I, 407 (tr. 491): at I, 417 (tr. 516-17), Juzjani refers to the conquest of Hindustan as far east as the borders of Ujjain (Habibi s edn reads

CYN in error), i.e. Malwa.

Machchlishahr copper-plate inscription, dated Vikrama samvat 1253/1197: P. Prasad, Sanskrit inscriptions,

58-70 (no. 11:6). Niyogi, History of the Gahadavala dynasty, 113ff. TN, I, 426-7 (tr. 558). The name

'Bangala' is not found prior to BaranI: Ahmad Hasan Dani, 'Shamsuddin Ilyas Shah, Shah-i Bangalah', in

Hari Ram Gupta et al. (eds.), Essays presented to Sir Jadunath Sarkar, Sir Jadunath Sarkar

commemoration volume, II (Hoshiarpur, 1958), 50-8.

~ disentangle plausibility from the hyperbole of our sources. At one extreme stand the enthusiastic claims

of Fakhr-i Mudabbir:

Infidel towns have become cities of Islam. In place of images, they worship the Most High. Idol temples

have become mosques, colleges (madrasaha) and hospices (khanaqahha). Every year several thousand

infidel men and women are being brought to Islam .. .65

We might be more inclined to accept Hasan-i Nizami's statements that Aybeg 'uprooted idolatry' and

'destroyed temples' at Kuhram, and that at Mirat, Banaras (a thousand temples here) and Kalinjar idol

temples were converted into mosques.66 Such thorough-going tactics are conceivable as far as the

respective urban centres were concerned. In some cases, too, architectural remains endorse Hasan-i

Nizami's claim that the stone from demolished Hindu temples was used in the erection of mosques, as for

example at Delhi and for the Arhai Din ke Jhompra mosque at Ajmer.67 But other assertions elicit a greater

degree of scepticism: that Aybeg freed the whole region (diyar) of Kol, rather than just the town, from idols and idol worship is doubtful.68 Moreover, the treatment of Hindu temples by the eighth-century Muslim

conquerors of Sind had varied with the circumstances,69 and we might reasonably assume that this was true

of the early thirteenth century also. Cities which capitulated - as for instance did Gwaliyor in 597/1200-1 -

presumably obtained a better deal for their temples than did places which had to be taken by storm.

Whatever the Muslim literati wanted people to think, the hallmark of these years was not uncompromising

iconoclasm.

The language of our sources has served to distort the character of these and later campaigns, so that they

have taken on the hue of a conflict that was religiously inspired - a development in turn nurtured by more

modern communalistic attitudes.70 For Juzjani, Mu'izz al-Din is always 'the holy warrior sultan' (sultan-i

ghazi), and Muslim writers designate his forces as 'the army of Islam'. When recounting the Ghurid

triumphs over the Indian infidel, Juzjani likens them to the victories of Mu'izz al-Din's contemporary Saladin over the Christian Franks- of Syria and Palestine.71 Yet it is important not to overstate the

significance of holy war in the Ghurid campaigns, at least as far as the sultan's motives were concerned.

Booty, to pay for the conflict with the Khwarazmshah, was undoubtedly a major

65 SA, 26. 66 Taj, fols. 53a, 74b, 134b, 185a.

67 Ibid., fols. 48a, 114. See Robert Hillenbrand, 'Political syrtibolism in early Indo-Islamic mosque

architecture: the case of Ajmir', Iran 26 (1988), 105-17.

68 Taj, fo\. 138a.

69 Friedmann, 'A contribution', 328-9; idem, 'The temple of Multan: a note on early Muslim attitudes to

idolatry', IOS 2 (1972), 176-82.

70

For a judicious treatment of this theme, see Carl W. Ernst, Eternal garden, Myyssticism, history, and

politics at a South Asian sufi center (Albany, New York, 1992), 18-29 passim.

71 TN, I, 290 (tr. 214).

~ incentive; and the distribution of find-spots for the coins minted in India by the conquerors is significant,

showing that a good proportion found their way back to the homeland.72 Muslim authors make great play of the golden artefacts from Ajmer which formed part of the Chawhan tribute, were forwarded to Ghiyath al-

Din Muhammad b. Sam and came to decorate the royal palace at Firuzkuh and the congregational mosque

at Herat.73 Both his successful attack on Daybul and his ill-fated Nahrwala campaign surely represent bids

by Mu'izz al-Din to restock his treasury by looting regions whose princes enjoyed a notoriously large

income from the proceeds of commerce; and there can be little doubt that the Daybul expedition, which

yielded great quantities of plunder,74 lubricated his subsequent war efforts against the Ghaznawids. Juzjani

would later hear from Mucizz al-Din's treasurer extraordinary figures for the weight of the gems obtained in

plunder from India and stored at Ghazna at the time of the sultan's death.75

Nor was the long-drawn conflict that marked the advance of Muslim power necessarily one that simply

pitted Hindu troops against Muslims. In the final assault on his co-religionist Khusraw Malik, Mu'izz al-

|Din had cooperated with the Hindu prince of Jammu, while the Ghaznawid Sultan had in turn been allied with the infidel Khokhars of the Panjab.76 We do not know at what point Mu'izz al-Din and his generals

followed the Ghaznawid example in employing contingents of Hindu troops. Aybeg's army at the siege of

Mirat certainly included Hindu soldiers; and when he advanced on Lahore in 602/1206, the 'Hindustan

forces' (hasham-i Hindustan) that accompanied him contained, we are told, 'ranas and thakurs' -Hindu

chiefs at the head of their own retinues, in the service of the Muslim warlord.77

For all these qualifications, however, the Delhi Sultanate was firmly rooted in a long tradition of Muslim

military activity within the subcontinent, and its rulers could be excused for seeing themselves as the latest

in a line of Muslim holy warriors. Pride of place among these undoubtedly went to Mahmud of Ghazna. It

is no accident that in his Fatawa-yi Jahdndari, a mirror for princes, the mid-fourteenth-century author

Barani produced what purported to be a political testament from Mahmud to posterity; or that his contemporary cIsami, modelling his epic Futuh al-Salatin on Firdawsi's Shah-Nama (which had been

dedicated to Mahmud), chose effectively to begin the work with Mahmud's own campaigns, and credited

him with the establishment of Islam in the subcontinent; or that Shams-i Siraj cAfif, describing Sultan Firuz

Shah's iconoclastic activities in Jajnagar

72 Deyell, Living without silver, 195, 203-6. 73 TN, I, 375 (tr. 404). SA, 22-3. Taj, fol. 80b.

74 TN, I, 397 (tr. 451-3). 75 Ibid., I, 404 (tr. 487-8).

76 Ibid., I, 398 (tr. 454-5). Bosworth, Later Ghaznavids, 129-30.

7 Mirat: Taj, fol. 74a. Lahore: SA, 33 (reading RANGAN for the RATGAN of the text). For Indian troops in the Ghaznawid armies, see Bosworth, Ghaznavids, 110; also verses attributed to the Ghurid cAla' al-Din

Husayn, cited in TN, I, 346 (tr. 357).

~ (Orissa) in 762/1361, likens him to Mahmud.78 Admittedly the great Ghaznawid 'amir was sui generis.

'The Almighty', wrote Juzjani, 'had conferred upon that ruler many superior characteristics ('alamat) and

miraculous" signs (karamat), which in their number and magnificence have not been combined since in any

other sovereign.'79 But this hardly rendered Mahmud any less worthy of emulation; the Delhi Sultans had

no more distinguished ideological forebear.80

78 FS, 28-9, 30, 609 (tr. 66-7, 68, 907). Shams-i Siraj cAfTf, Ta'rlkh-i Firuz-Shahl, ed. Maulavi Vilayat

Hosain (Calcutta, 1888-91), 170. 79 TN, I, 230 (tr. 83 modified); cf. also I, 229 (tr. 77-80).

80

Nizami, Some aspects, 107-9, and On history and historians, 107-9; C. E. Bosworth, 'Mahmud of Ghazna

in contemporary eyes and in later Persian literature', Iran 4 (1966), 89-90, repr. in his Medieval history.

Aziz Ahmad, Studies, 79.

~Map 1: The eastern Islamic world in 1206

~CHAPTER 2

From Ghurid province to Delhi Sultanate

Ghurid government

The conquerors brought with them the institutions to which they were accustomed in the Ghurid

homelands. Chief among these was the iqta' (frequently and misleadingly rendered as 'fief'), the

transferable revenue assignment in lieu of salary for service (usually military service), which by 1200

already had a long history in the eastern Islamic territories, having been adopted by the Ghaznawids and

having reached its highest expression under the Seljiiks; imported into India by the Ghurids, it would form

one of the characteristic institutions of the early Delhi Sultanate. Various developments had occurred in

twelfth-century Persia to blur the nature of the iqta and to assimilate it to an administrative command.1 Hence our sources, in the terminology they employ for the grants made by Mucizz al-Din, are often less

than helpful. Aybeg's earliest assignment, at Kuhram, is described by Juzjani as iqta', whereas Fakhr-i

Mudabbir speaks of it simply as the 'command' (sipahsalari) there and Hasan-i Nizami says that he was

given the governorship (ayalat) of Kuhram and Samana.2 Yet we are left in no doubt that the iqta' was

widespread in northern India by the time of the creation of the independent Delhi Sultanate. Iltutmish

became iqta'-holder (muqta1) of Baran under Ay beg; and we find iqtacs in Awadh before 1200 and in west

Bengal following its conquest by Muhammad b. Bakhtiyar. The term was also used of the holdings of

ordinary troopers: Ibn al-Athir refers to such men as holding iqta's from Aybeg in 602/1206.3

1 See Claude Cahen, 'L'evolution de l'iqta0 du ixe au xiiie siecle', Annales, Economies, Socie'te's,

Civilisations 8 (1953), 25-52, repr. in his Les peuples musulmans dans I'histoire medievale (Damascus, 1977), 231-69; idem, 'Ikta0', Enc.M2; A. K. S. Lambton, 'Reflections on the iqta", in G. Makdisi (ed.),

Arabic and Islamic Studies in honor of Hamilton A.R. Gibb (Leiden, 1965), 358-76; eadem, Continuity and

change in medieval Persia (London, 1988), 99-113. There is a succinct account in D.O. Morgan, Medieval

Persia 1040-1797 (London, 1988), 37-40.

2 Taj, fol. 51b. TN, I, 417 (tr. 515).

3 IA, XII, 140/214. Bengal: TN, I, 422, 423, 432, 433 (tr. 549-50, 572, 574, 575). Baran: ibid., I, 443, and

II, 19 (tr. 604, 745).

~ As far as we can tell from the exiguous material in our sources, the hierarchy of Ghurid officials at

Flruzkuh and Ghazna did not differ appreciably in its outlines from those maintained by other eastern Islamic dynasties. The wazir ('minister'), as elswhere, headed the civil administration at Ghazna; we also

read of the treasurer (khazin) and the overseer of public morality/inspector of the markets (muhtasib). The

appointment of judges (quddat, sing, qadi) who enforced the religious law, the SharFa, was also in the

sultan's hands. The army had its own SharFa court under its own judge (qdai-yi lashgar), though the two

offices could evidently be combined.4 It is possible to draw too sharp a line between the civil and the

military. Mu'ayyad al-Mulk Sajzi, who served first Mu'izz al-Din and then Yildiz as wazir at Ghazna, also

acted on occasions as a military commander, as would the wazirs of the early Delhi Sultans; and we find a

contingent (khayl) of horsemen from Tulak in Mu'izz al-Din's service led by its qadi, Juzjani's kinsman

Diya' al-DIn.5 The distinction between 'men of the sword' and 'men of the pen' (arbab-i tigh-u qalam), to

borrow the widespread term used by our sources,6 or that between Turkish military and Persian ('Tajik')

bureaucrats, was evidently in practice sometimes rather blurred.

Most of the offices of which we read are essentially military: the commander of the sultan's guards or

executioners (sar-i jdndar); the chief armour-bearer (sar-i sildhdar); the muster-master ('arid), who seems

to have performed the functions of a minister of war; the military chamberlain (amir-hdjib), often entrusted

with command in the field; the military justiciar (amir-i dad), who at Ghazna, if Ibn al-Athir is to be

believed, commanded the citadel; the intendant of the sultan's stables (amir-i akhur), an office held by Qutb

al-Din Aybeg himself prior to his appointment in India; and the intendant of the hunt (amir-i shikar), a

position of some importance under a regime in which the chase constituted both the monarch's chief

recreation and a valuable form of military exercise for the troops.7

Our knowledge of the administration of Mu'izz al-Din's conquests in northern India is patchy. Aybeg

clearly had his own staff which mirrored that of the sultan at Ghazna, including a sar-i jandar and an amir-i

shikar;8 and there was an 'arid at Delhi, of whom we know only that he rejected Muhammad b. Bakhtiyar

as unfit for service and thereby unwittingly launched him on a more dazzling career.9 At certain centres, the

local

4 Ibid., I, 367, 389,405 (tr. 389, 430, 489).

5 Ibid., I, 380-1, 419 (tr. 415, 534), for Mu'ayyad al-Mulk; I, 398-9, 400 (tr. 457-8, 464), for the qadi of

Tulak. On Iltutmish's wazir Junaydi, see below, p. 35.

6 Taj, fols. 106b, 135b, 218a; and cf. AH, 138.

For these offices, see generally Heribert Horst, Die Staatsverwaltung der Grosselguken und Uorazmsahs

(Wiesbaden, 1964), passim. Amir-i dad at Ghazna: IA, XII, 143/217, 145/221. Aybeg as amir-i akhur. TN,

I, 303, 416 (tr. 248, 514). On the passion of the Ghurid Sultan Ghiyath al-Din Muhammad b. Sam for the

chase, see ibid, I, 364-5 (tr. 385-7). 8 Ibid., I, 443 (tr. 603). 9 Ibid., I, 422 (tr. 549).

~ military justiciar (amir-i dad) seems to have enjoyed a pivotal position; perhaps, like his counterpart at

Ghazna, he was in command of the citadel. Examples are the officer at Multan, treacherously killed at the

onset of a rebellion in 601/1204, and cAli-yi Ismail at Delhi, instigator of the coup that conferred power on

Iltutmish in 607/1210-11.10

The emergence of an autonomous Muslim power in India

The events that followed Mucizz al-Din's death represented a disjunction from previous developments.

Hitherto Delhi had been merely one of the Muslims' forward bases, and Lahore had remained the capital of

Mucizz al-Din's Indian province just as it had been of the Ghaznawid territories in India. For Fakhr-i

Mudabbir, Lahore was 'the centre of Islam in India' (markaz-i Islam-i Hind); while Hasan-i Nizami,

describing how the city was conferred on Iltutmish's eldest son in 614/1217, could still observe, wistfully

perhaps, how Lahore had 'ever been the residence of celebrated maliks and the seat of powerful rulers'.11

Within a few decades of Iltutmish's seizure of power at Delhi and the creation of the independent Sultanate,

the steady build-up of Mongol pressure made Delhi appear a far more suitable residence for its rulers than

was Lahore, a circumstance incidentally emphasized by the Mongols themselves when they took and

sacked Lahore in 639/1241. But that Delhi had become the capital of Muslim India and the seat of independent monarchs was in some measure a historical accident, though our principal sources are by no

means anxious to acknowledge it.

In view of his subsequent rise, it is easily forgotten that when in 588/1192 Qutb al-Din Aybeg was

stationed at Kuhram he was one of Mucizz al-Din's more junior slaves, in contrast, for example, with Taj al-

Din Yildiz, who seems to have been among the most senior and is said to have been made their commander

(sarwar).12 Aybeg's precise status within the Ghurid conquests is obscure. We find him appointing amirs to

certain strongpoints: his own ghulams Iltutmish first to Gwaliyor and later to Baran and Bada'un, and

Aybeg-i Tamghaj to Tabarhindh; and other Turkish officers like Hasan-i Arnab to Kalinjar and Husam al-

Din Oghulbeg to Kol.13 The Khalaj freebooter Muhammad b. Bakhtiyar waited upon Aybeg at Bada'un

10 Taj, fols. 188a, 189a. TN, I, 444 (tr. 605).

11 SA, 30. Taj, fol. 259a; see also fol. 211a, mustaqarr-i sarlr-i salatin.

12 TN, I, 411 (tr. 498); cf. also IA, XIJ, 141/215. A later tradition made Yildiz Mucizz al-Din's adopted son:

TFS, 550. Both names are Turkish: for yildizl'yulduz, 'star', see Sir Gerard Clauson, An etymological

dictionary of pre-thirteenth-century Turkish (Oxford, 1972), 922-3; for ay, 'moon', and beg, 'prince', 'lord',

Jean Sauvaget, 'Noms et surnoms de Mamelouks', JA238(1950), 31-58 (no. 37).

13 TN, I, 443 (tr. 603-4). Taj, fols. 138a, for Oghulbeg, and 185a, for Hasan-i Arnab. The second element in the latter name, usually read as 'Arnal', is clearly 'RNB in the best mss.; see also Hodivala, Studies, II, 53-4.

Oghulbeg's name is Tu. oghul, 'son', 'boy' (Clauson, Etymological dictionary, 83-4), + beg, 'prince'.

~and accepted from him the privilege of maintaining a band (nawbat) outside his residence, together with a

kettle-drum and standard and a diploma that confirmed and, according to Hasan-i Nizami, extended the

territory under his control.14 Such attentions indicate that he acknowledged Aybeg's authority as the sultan's

representative.

Yet there clearly existed other commanders in India who were independent of Ay beg. It must be

remembered that the sources tell us only of the operations conducted by Mu'izz al-Din and by Aybeg as his

lieutenant, together with (in the case of Juzjanl and Ibn al-Athir) the activities of the Khalaj forces in Bihar

and Bengal. Of other campaigns, led by other Ghurid officers, which must have taken place in these years, we learn little. Baha' al-Din Toghril, a senior ghulam of the sultan, had received from Mucizz al-Din in

person his command at Thangir and the task of reducing the great fortress of Gwaliyor, so that he deeply

resented the surrender of Gwaliyor to Aybeg in 597/1200-1. His inscriptions suggest that he proclaimed

himself sultan at some point following Mucizz al-Din's murder.15 We do not know in what relationship

Aybeg stood to cIzz al-Din 'All at Nagawr or to Nasir al-Din Aytemur at Uchch.16 Nor are we told whether

prior to 602/1206 his writ extended to Lahore, which seems to have constituted a joint command with

Multan.17 It is indeed possible that he wielded no authority in those tracts which had formed part of the

Ghaznawid state at the time of its conquest in 582/1186.

The evidence does not, in other words, sustain the belief of modern historians that Aybeg was left as the

sultan's deputy in the Indian provinces in the wake of the Tara'in victory: he is admittedly so designated by Ibn al-Athir, but only in the context of the events following the sultan's assassination.18 Earlier the same

author describes him simply as commander of the sultan's forces in India.19 Even on the testimony of his

panegyrist Fakhr-i Mudabbir (writing soon after Aybeg's assumption of the royal dignity at Lahore), his

promotion to the status of viceroy of the entire province 'from the gates of Peshawar to the furthest parts of

India' had occurred only a few weeks prior to Mucizz al-DIn's death.20

14 Taj, fol. 186. TN, I, 423-7 (tr. 551-4, 560).

15 Mehrdad and Natalie H. Shokoohy, 'The architecture of Baha' al-Din Tughrul in the region of Bayana,

Rajasthan', Muqarnas 4 (1987), 114-32 (esp. 115); see also Mehrdad Shokoohy, Rajasthan I (London,

1986), 51-3; TN, I, 421-2 (tr. 544-7), for his earlier career. The name is Tu. toghril, 'bird of prey': Clauson,

Etymological dictionary, 472; TMENP, III, 346-8 (no. 1345). 16 TN, I, 419 (tr. 531-2), for Aytemiir ('YTM in Habibi's edition): ay, 'moon', and temur, 'iron', see Sauvaget,

'Noms et surnoms', no. 42. The reference to cAli-yi Nagawri is omitted in Habibi's text (I, 422), but cf. his

apparatus and Raverty's tr., 549.

17 IA, XII, 136/209, 137/210; and cf. TN, I, 398 (tr. 456). I cannot agree with I. Habib, 'Formation', 6, that

Aybeg had charge of Lahore prior to 1206.

18 IA, XII, 164/248. Habibullah, Foundation, 63, wrongly describes Delhi as 'the capital of Muizzuddin's

Indian dominion'.

19 IA, XII, 136/209, muqaddam 'asakiri'l-Hind. 20 SA, 28.

~ Whatever his position when the sultan was murdered, Aybeg was able to move from Delhi to Lahore and

to take up his quarters as ruler there on 18 Dhu'l-Qa'da 602/26 June 1206. Juzjani at one point claims that

he did so following the arrival of a diploma from the new sultan, Mu'izz al-Din's nephew Ghiyath al-Din

Mahmud b. Muhammad.21 Yet this is unlikely on chronological grounds. More probably Aybeg was

encouraged to do so by a sudden access of manpower, since soon after Mucizz al-Din's assassination his

wazir sent back to India all those troopers currently with the late monarch's army who held iqta's from

Aybeg.22

Nevertheless, in contrast with Yildiz, who at Ghazna ignored the rights of the new Ghurid

sovereign, Aybeg maintained the khutba for Ghiyath al-Din Mahmud and struck coins in his name, gestures that were rewarded in 605/1208-9 with the gift of a ceremonial parasol (chatr) and, allegedly, with the style

of sultan.23 He is called sultan by Hasan-i Nizami, though no coins of his have come down to us bearing

that title.24

Aybeg also secured recognition of his authority in Bengal after the assassination of Muhammad b.

Bakhtiyar by one of his Khalaj officers, 'Ali-yi Mardan, in c. 602/1205-6. He first sent an army from

Awadh under Qaymaz-i Rumi, who defeated and killed the new Khalaj ruler, Muhammad-i Shiran, and

installed at Deokot, one of the principal Muslim-held towns, another of Muhammad b. Bakhtiyar's Khalaj

lieutenants, Husam al-DIn Twad. Subsequently Aybeg conferred a robe of honour on 'Ali-yi Mardan, who

had taken refuge at Lahore, and despatched him eastwards as his subordinate. On his arrival at Deokot,

'Ali-yi Mardan was received submissively by Twad, and established himself as ruler of the entire Muslim

territory in Bengal.25 Thus for most of his reign Aybeg was represented in the east by his own nominee, and had emerged as the paramount ruler in Muslim India.

Aybeg's most dangerous rival was Taj al-Din Yildiz, with whom he engaged in a struggle for possession of

Ghazna, first inciting one of Yildiz's officers to seize the place in 603/1207 and briefly occupying Ghazna

in person two years later. Juzjani's description of this conflict as originating over Lahore suggests that

Yildiz claimed to rule all Mu'izz al-Din's eastern territories.26 Their rivalry seems to have kept Aybeg at

Lahore during his

21 TN, I, 417 (tr. 521-5). 22 IA, XII, 140/214.

23 TN, I, 373 (Ghur and Ghazna, erroneously included among the territories covered by the mandate in Habibi's text, are omitted in BL m., fol. 152a, and Raverty's tr. 398). IA, XII, 165/249.

24 Taj, fols. 211a, 217a. P. Jackson, 'Kutb al-DIn Aybak', Enc.Isl2.

25 TN, I, 432-3, 434 (tr. 572-6, 577-8). The name of Aybeg's general is Tu. qaymaz, 'he who does not turn

back': Sauvaget, 'Noras et surnoms', no. 150. For Deokot, at 25° 11 'N., 88° 31'E., see Hodivala, Studies, I,

209, and II, 57.

26 TN, I, 417 (tr. 526); I, 412, 413-14, 417 (tr. 503, 506, 526-8), for the later struggle over Ghazna. For the

earlier episode, in 603/1206-7, see I A, XII, 165-6/249-50. See generally Jackson, Fall of the Ghurid

dynasty'.

~ four-year reign, either guarding against an invasion of the Panjab or seeking yet another opportunity to

take Ghazna. The dearth of evidence for military operations against independent Hindu states at this time

presents a marked contrast with the era of his lieutenancy on behalf of Mu'izz al-Din.27

When Aybeg died in a polo (chawgan) accident in 607/1210-11, his ghulam Iltutmish was invited into

Delhi from Bada'un by a party in the city, headed by the military justiciar (ami-i dad) cAli-yi Ismail, and set

himself up as ruler. A certain obscurity surrounds his rival Aram Shah, who at one point is called Aybeg's

son.28 He seems, however, to have reacted sharply to Iltutmish's seizure of power at Delhi. His supporters

included some of the Mucizzi amirs, i.e. former officers of Mucizz al-Din, who left Delhi to join the

opposition to Iltutmish; the latter, on the other hand, had the backing of the Qutbi amirs, namely the

servitors of Aybeg. Hasan-i Nizami, who ignores Aram Shah and for whom all this is simply a rising against his patron Iltutmish by a group of recalcitrant Turks, names as their leader the sar-ijandar, *Berki.

They advanced from Lahore to Delhi, where Iltutmish met and defeated them in the Bagh-i Jud. Aram Shah

is alleged to have been 'martyred'; but whether he was killed in the engagement or put to death as a

prisoner, we are not told.29

Shams al-Din Iltutmish30 was at first only one of a number of Muslim rulers in the subcontinent, and his

position was highly precarious, even after the overthrow of Aram Shah. In the following century it was

remembered that he had been obliged to dispute his new kingdom with amirs who held iqta's in 'Hindustan'

by grant from Mu'izz al-Din.31

Juzjani speaks of campaigns by which Iltutmish brought under his control

'the outlying regions which were dependent on Delhi' (atraf-i mamalik-i madafat-i hadrat-i Dihli), singling out for particular mention Bada'un, Awadh, Banaras and the Siwalik; elsewhere he alludes briefly, in his

list of Iltutmish's conquests, to the capture of Banaras and the flight of Qaymaz, presumably Aybeg's

former officer whom we have already encountered.32

Further afield, Iltutmish could command no allegiance whatever. On the news of Aybeg's death, his client

in Bengal, cAH-yi Mardan, assumed

27 P. Hardy, 'Dihli Sultanate', Enc.Isl.2, Ill, 269a. Jackson, 'Kutb al-DIn Aybak'.

28 Only in a chapter heading: TN, I, 418 (tr. 528).

29 Ibid., I, 418, has simply Aram Shahra qada-yi ajal dar rasid; but cf. BL ms., fol. 168b, Aram Shahra ... shahld kardand (also Raverty's tr. 530). The 'revolt' is described in TN, I, 444 (for the translation, which is

garbled by Raverty, see I. Habib, 'Formation', 9 and n.50), and in Taj, fols. 219b-224a. Modern authors

have usually rendered the sar-i jandar% name as Turki', but the mss. read .YRKY and TYRKY.The name is

possibly connected with Tu. berk, 'firm', 'solid': Clauson, Etymological dictionary, 361-2.

30 The correct form of his name has been established by Simon Digby, 'Iletmish or Iltutmish? A

reconsideration of the name of the Dehli Sultan', Iran 8 (1970), 57-64.

31 TFS, 550.

32 Conquests: TN, I, 444-5 (tr. 607-8). Qaymaz: I, 452 (to be corrected from BL ms., fol. 179b; cf. also Raverty's tr., 627).

~ overeign status, entitling himself Sultan cAla' al-Din.33 In Sind, Multan was seized by Nasir al-Din

Qubacha, a former ghulam of Mu'izz al-Din w'ho had been muqta' of Uchch since 601/1204 and who now

proclaimed his undependence with the adoption of two chatrs. Lahore was disputed between Qubacha and

Yildiz.34 Iltutmish secured his position at first by icknowledging Yildiz's sovereignty, receiving in return

the insignia of royal power, a chatr and a durbash or ceremonial baton. His earliest inscription, dated

Jumada I 608/October 1211, styles him not sultan but only king (al-Malik al-Mu'azzam)?5 We cannot fail

to be struck, again, by the relative absence of campaigns against the Hindu powers. During the first sixteen

years of his reign, Iltutmish is known to have conducted only one such expedition, against the Chawhan

ruler of Jalor, which is described by Hasan-i Nizami but is undated; it may have occurred not long after the

suppression of Aram Shah's attempt on Delhi. That other major undertakings were deferred until the attacks on Ranthanbor (623/1226) and Mandor (624/1227) was clearly due to pressing concerns elsewhere.

Iltutmish's first opportunity came in 612/1215-16, when Yildiz was forced out of Ghazna by the

Khwarazmshah and fell back on the Panjab. Having wrested Lahore from Qubacha, he then pushed into

Delhi territory in an effort to make good his rights over his subordinate Iltutmish, issuing demands of the

kind, says Hasan-i Nizami, that no sovereign could demean himself to answer. Iltutmish met him on 3

Shawwal 612/25 January 1216 on the historic battlefield of Tara'in, near Samana: Yildiz was defeated and

incarcerated in Bada'un, where he was later put to death.36 But if the elimination of Yildiz conferred

independence on the Delhi ruler, it did not immediately result in any significant addition to his territory.

Lahore was reoccupied by Qubacha, whose empire now stretched from the Arabian Sea and the Indus delta

as far north as Nandana and Peshawar. Iltutmish seized Aybeg's old capital late in 613/in the winter of 1216-17, conferring it on his eldest son Nasir al-DIn Mahmud; but it continued to change hands thereafter,

while the two rulers disputed possession of Tabarhindh, Kuhram and Sarsati in the eastern Panjab.37 And

that Qubacha did not lack partisans in Delhi territory emerges from the account of Baha" al-Din 'A1i b.

Ahmad

33 Ibid., I, 434 (tr. 578). M. Nizamuddin, 'A rare coin of Ali Mardan Khalji - a medieval Muslim ruler of

Bengal', JNSI49 (1987), 50-5.

34

TN, I, 418, 419 (tr. 530, 532); for Yildiz at Lahore, see I, 444 (tr. 607). Qubacha's appointment to Uchch

dated from the death of its muqtac, Aytemur, in battle at Andkhud. His name is a derivative of Tu. quba, 'pale', 'pale yellow': Clauson, Etymological dictionary, 581.

35 M. M. Shu'aib, 'Inscriptions from Palwal', E1M (1911-12), 2-3; RCEA, X (Cairo, 1939), 72-3 (no. 3703).

TN, I, 444 (tr. 607).

36 IA, XII, 203/311-12. TN, I, 413, 445 (tr. 505, 608). Taj, fols. 238b-247b, with the precise date.

37 TN, I, 419, 444 (tr. 534, 607). Taj, fols. 253a-259b, for Iltutmish's first occupation of Lahore. But at the

advent of the Khwarazmians (below), it was held by Qubacha's son. Nandana: Nasawi, Sir at al-Sultan

Jalal al-Din, ed. O. Houdas (Paris, 1891), 86, and tr.

~ Jamaji in 'Awfi's biographical dictionary Lubab al-Albab. Jamajl, whom Iltutmish had put in command of Bahraich, declared for Qubacha and in 617/1220 sent to Uchch offering his submission. Writing at this very

moment, 'Awfi is unaware of the sequel (as, regrettably, are we), though he expresses the hope that his

master Qubacha will soon acquire dominion over the whole of Hindustan.38 The episode suggests that

Qubacha may have constituted a formidable threat to the Delhi ruler on the eve of the Khwarazmian

invasion.

As usurpers, Aybeg and Iltutmish stood in need of legitimation, and obtained it, on one level, from their

panegyrists. The sources for this era, with the exception of Ibn al-Athir, all emanate from within India.

They also date from the period following Mu'izz al-Din's death, and their accounts read as though there

were a continuity between their new masters and the Ghurid dynasty. Thus Fakhr-i Mudabbir - writing, it

will be recalled, at Aybeg's court soon after 602/1206 - has Aybeg taking over at Lahore in that year by virtue of his status as Mu'izz al-Din's deputy (qa'im-i maqam-u wali-'ahd) throughout his Indian

dominions.39 Juzjani, who composed his Jabaqdt-i Nasiri in Delhi when the Sultanate had been in existence

for five decades, clearly sought to gloss over Iltutmish's struggles with other former Ghurid lieutenants and

to portray him and his dynasty likewise as Mu'izz al-Din's true successors in India. Hence in the

introduction to his section on the Ghurids, the current sultan of Delhi, Nasir al-Din Mahmud Shah b.

Iltutmish, becomes 'the heir to that sovereignty and duly appointed successor (qa'im-i ma'mur) to that

kingdom'.40 That this is not intended merely as a figure of speech is clear from what Juzjani says later.

Approached by a courtier, who lamented the sultan's lack of sons to inherit his dominions, Mu'izz al-DIn

replied:

'Let other sultans have one son or two. I have several thousand sons - Turkish slaves - whose inheritance

will be my kingdom: after me, they will maintain the khutba in [my] empire in my name.' And it transpired as that ghazi monarch pronounced. Since his time, right down until these lines are being written, they have

preserved the whole empire of Hindustan and are still preserving it.. .41

Although some modern commentators have adopted this perspective on

Z. M. Buniiatov, Zhizneopisanie Sultana Dzhalal ad-Dina Mankburny (Baku, 1973), 131. Peshawar: TJG,

I, 61 (tr. Boyle, 328).

38 'Awfi, Lubab al-Albab, ed. E. G. Browne and M. M. Qazwini (Leiden and London, 1903-6, 2 vols.), I,

115. The summary of Jamaji's career in Nizamu'd-din, Introduction, 13-14, is inaccurate.

39 SA, 28; cf. also 33, in wall-cahdi.

40 TN, I, 323 (tr. 310 modified). Peter Hardy, 'Force and violence in Indo-Persian writing on history and

government in medieval South Asia', in Milton Israel and N. K. Wagle (eds.), Islamic society and culture.

Essays in honour of Professor Aziz Ahmad (Delhi, 1983), 180-1.

41 TN, I, 410-11 (tr. 497-8 modified); cf. I, 415 (tr. 508-12), where the Mu'izzi sultans, i.e. Aybeg and

Yildiz and their contemporaries, are described as the late sovereign's heirs, and I, 393 (tr. 438), where

Mu'izz al-Din is said to have entrusted (sipurd) Ghazna to Yildiz.

~ events,42 it is in fact highly dubious and smacks of ex post facto justification: in Islamic law it was the

master who inherited from the slave and not vice versa.431 But evidently the sovereignty of the early Delhi

Sultans, in the eyes of certain of the Muslim intelligentsia, required the sanction of Mu'izz al-Din.

The Khwarazmian and Mongol invasions

In Iltutmish's early years, there was no guarantee that the former territories of the Ghurids in Sind and the

Panjab would not share the fate of the regions beyond the Indus and be absorbed into the Khwarazmian

empire. Following his seizure of Ghazna, the Khwarazmshah 'Ala' al-Din Muhammad created for his son

Jalal al-DIn *MingbarnI a large appanage that stretched as far as the Indus. At some point his forces

wrested Peshawar from Qubacha.44 A casual observer might have been forgiven for supposing that the

Ghurids' erstwhile lieutenants were similarly destined for political oblivion. It was fortunate for the fledgeling Delhi Sultanate, as indeed for the other powers which had inherited the Ghurid mantle in

northern India, that when the Khwarazmians appeared in force in the Panjab they came not as conquerors

supported by the resources of an extensive Central Asian empire, but as fugitives. At the height of his

power, the Khwarazmshah clashed with his new and formidable neighbours, the Mongols led by Chinggis

Khan, whose great westward campaign of 615-622/1218-1225 destroyed the Khwarazmian polity and

devastated eastern Persia. Muhammad died in misery on the coast of the Caspian Sea in 618/1221.45 At an

early stage in this crisis, it seems, India was being advocated as a refuge for the dynasty, possibly because

its rulers were deemed inadequate to resist.46 Some of the Khwarazmshah's advisers urged him to make a

stand at Ghazna and to retreat into India if this failed; their counsel was rejected.47 In the event, it was the

shah's son and effective successor, Jalal al-Din, who entered India following a crushing defeat by Chinggis

Khan on the banks of the Indus in Rajab 618/November 1221 and began a career of aggression in northern India that was to last for almost three years.

42 Aziz Ahmad, Political history, 99; see also 6, 84, 97, 118, 146, 149 (though at 13 he is prepared to

consider the possibility that Mu'izz al-Din's dominions were misappropriated).

43 Patricia Crone, Roman, provincial and Islamic law (Cambridge, 1987), 36-8. P. G. Forand. 'The relation

of the slave and the client to the master or patron in medieval Islam', International Journal of Middle East

Studies 2 (1971), 61.

44 P. Jackson, 'Jalal al-Din, the Mongols and the Khwarazmian conquest of the Panjab and Sind', Iran 28

(1990), 48; the form ('Mingirini') adopted there for the prince's name must now be discarded on the basis of

numismatic evidence (unpublished paper by Mr William Spengler). 45 Barthold, Turkestan, 393-439, 446 ff. J. A. Boyle, 'Dynastic and political history of the II-khans', in

Boyle (ed.), Cambridge history of Iran, V, 303-17.

46 See, for instance, the advice later given to Jalal al-Din in India by one of his lieutenants, as reported by

Nasawi, 91 (tr. Buniiatov, 136).

47 TJG, II, 106 (tr. Boyle, 376).

~ Juzjani, regrettably, says very little about the Khwarazmian invasion of the Panjab, perhaps because the

episode reflected no credit on Iltutmish, who failed to collaborate with a fellow-Muslim against the pagan Mongols. Ibn al-Athir's information on India, too, seems suddenly to dry up after the downfall of Yildiz.

Jalal al-Din's exile in the Panjab and Sind is dismissed in a couple of lines; he is lost to sight until his

emergence from the Makran desert in 620/1223.48 We are therefore fortunate that the prince's biographer

Nasawi, writing in 639/1242, and the later writer Juwayni, whose history of the Mongols dates from c.

658/1260, supply numerous (if sometimes conflicting) details which enable us tentatively to reconstruct

events.49 Jalal al-Din was soon reinforced by fresh refugees from Persia and made war upon Qubacha, who

was defeated near Uchch and obliged to become tributary to the Khwarazmians: his son, who had rebelled

against him at Lahore, likewise yielded and became Jalal al-Din's lieutenant there. Subsequently, when

Qubacha refused to assist him against the pursuing Mongols, Jalal al-Din ravaged the neighbourhood of

Uchch before moving south into the lower Indus region. Here he received the submission of Siwistan (now Sehvan), occupied Daybul, and despatched a plundering expedition to Nahrwala in Gujarat. At this juncture

he heard reports of the eagerness with which the subjects of his surviving brother in Persia desired his

return, and chose to make his way to western Persia by way of the Makran desert late in 620/1223, in the

hope of rebuilding his father's empire. He was eventually killed in Azerbaijan while in flight, again, from

the Mongols in 628/1231.

The Delhi Sultan was inevitably drawn into the conflicts between his neighbours and the Khwarazmians.

When Jalal al-Din at some stage advanced to within a few days' journey of Delhi, requesting asylum and

proposing an alliance against the Mongols, Iltutmish - not unnaturally reluctant to jeopardize his new-found

autonomy by installing the Khwar-azmshah close at hand - had the envoy murdered and returned an

evasive answer. Iltutmish then sent troops to aid Qubacha prior to the battle near Uchch, and subsequently

moved against Jalal al-Din in person: on this occasion the vanguards clashed but the two sovereigns exchanged friendly messages and withdrew. Whether Iltutmish, at least, was sincere is open to doubt. One

of the reasons underlying the Khwarazmshah's departure from India, according to Nasawi, was the news

that a coalition had been formed against him by Iltutmish, Qubacha and a number of Hindu chiefs (rayat

wa-takakirat, 'rais and thakurs'), who had occupied the banks of the Jajner river (presumably the Sutlej) in

order to cut off his retreat.

The Mongols who pursued the Khwarazmshah into India do not appear

48 IA, XII, 276/425-6.

49 For what follows, see Jackson, 'Jalal al-Din', which also discusses the sources, including the large anonymous fragment (hitherto little used), Bodleian ms. Th. Hyde 31. Brief biography of Jalal al-Din in J.

A. Boyle, 'Djalal al-Din Khwarazm-Shah', Enc.Isl2.

~ to have encroached upon Iltutmish's territory.50 Chinggis Khan, who briefly contemplated returning to

Mongolia by way of a more direct route through the Himalayan foothills, sent envoys to the sultan asking

his permission to move through his dominions. We are not told the fate of this embassy -simply that

Chinggis Khan abandoned his intention in the face of unfavouable auguries. Juwayni, however, alleges that

he advanced several stages before turning back because there was no way through, and only then withdrew

through Peshawar.51 His general Dorbei sacked Nandana, currently held by one of Jalal al-Din's lieutenants,

and in the late winter of 621/ 1224 laid siege to Multan, where a spirited defence was conducted by

Qubacha himself. But after an investment which according to Juzjani lasted for forty days (or three months

if we accept the testimony of another source), he abandoned the attempt in view of the onset of the hot weather. As they retreated, the Mongols ravaged the regions of Multan and Lahore. Almost another two

decades were to elapse before they entered the territory of the Delhi Sultanate.

Jalal al-Din's departure from India did not mean the end of the Khwarazmian occupation. His lieutenant,

Jahan-Pahlawan Ozbeg-bei, was entrusted with the Khwarazmshah's Indian conquests, while those parts of

Ghur and Ghazna which had not so far been ravaged by the Mongols were conferred upon Sayf al-Din

Hasan Qarluq, sumamed Wafa Malik.52 But as Mongol pressure was maintained by the forces left behind

by Chinggis Khan, so the residue of Khwarazmian dominion west of the Indus began to crumble. Ghur

seems finally to have been overrun in 623/1226, and in that same year Qubacha was called upon to repulse

a band of Khalaj tribesmen previously in the Khwarazmshah's service - arid presumably, therefore, under

the nominal authority of Hasan Qarluq - who had entrenched themselves in lower Sind.53

Iltutmish's conquest of Muslim India

Qubacha's empire in the Indus valley had in all probability been gravely weakened already prior to this

intrusion, for it had borne the brunt both of Jalal al-DIn's attacks and of the Mongol devastation. In the

event, however, it was to receive the coup de grace not from the north-west but from Delhi.

50 Mongol operations in India in 618-621/1222-4 are surveyed in Jackson, 'Jalal al-Din', 47-8, 50; see also

Paul Ratchnevsky, Genghis Khan: his life and legacy, tr. T. N. Haining (Oxford, 1991), 134. For the

general's name (Raverty's 'Turti'), see J. A. Boyle, 'The Mongol commanders in Afghanistan and India according to the Tabaqat-i Nasiri of Juzjani', 75 2(1963), 238-9.

51 TN, II, 126-7, 146 (tr. 1045-7, 1081-4). TJG, I, 109-10 (tr. Boyle, 137-9).

52 Nasawi, 92 (tr. Buniiatov, 136). Sayfi, Ta'r'ikh-Nama-yi Harat (c.1322), ed. M. Z. as-Siddiqui (Calcutta,

1944), 198, says that the rule of 'Malik Wafa' was later remembered in the Mastung region.

53 TN, 1,420 (tr. 539-41).

~ Both Hasan-i Nizami, speaking in the context of the Delhi forces' campaign against Lahore in 613/1216-

17, and Awfi, introducing Qubacha's overthrow twelve years later, allude to 'undertakings', 'promises' and

'treaties' of which Qubacha was unmindful.54 This suggests that the alliance with Iltutmish had its price. It is possible that in order to secure assistance from Delhi against the Khwarazmians - and perhaps also

against Yildiz earlier -Qubacha had either made some gesture in recognition of Iltutmish's sovereignty or

had promised to surrender territory to Delhi, and that his failure to abide by his obligations served as a

casus belli. There may well be some connection with a campaign by Iltutmish's forces to which Juzjani

refers obscurely, some years before the final conquest of Sind and resulting, apparently, in the capture of

Qubacha and the occupation of the district of Ganjrut (or Wanjrut) in the Multan province.55

Whatever the case, Qubacha had by 625/1228 finally lost the disputed regions of Tabarhindh, Kuhram and

Sarsati, which are found at that date in the hands of officers appointed by his rival. Lahore had also been

wrested from him, since Iltutmish's muqta there, Nasir al-Din Aytemur al-Baha'i, now presided over the

surrender of Multan.56 Another army appeared before Uchch, where it was soon joined by Iltutmish in person: the city capitulated at the end of Jumada I 625/early in May 1228. Qubacha had meanwhile fled to

the island stronghold of Bhakkar, pursued by a force under Iltutmish's wazir, Nizam al-Mulk Junaydi, and

on the night of 19 Jumada 11/26 May threw himself into the Indus to avoid being taken alive. From Daybul

Iltutmish received the submission of its prince, Chanisar, with the result that his sovereignty was

acknowledged as far as the Arabian Sea.57

The conquest of Sind left Iltutmish free to move against Jalal al-Din's

54 Taj, fol. 251b. JH, I, 10, and BL ms. Or. 2676, fol. 232a.

55 TN, II, 4 (tr. 723), with the impossible year 628 - perhaps an error for 618 or, given the similarity of

thaman"1 and thalath in the Arabic script, for 623. The printed text has WNJRWT, and Raverty identified the place with Bijnoot (Vijnot), on the fringes of the Bikaner desert and well to the south of Uchch; but BL

ms., fol. 197a, and IOL ms. 3745, fol. 268b, have KNJRWT. Ganjrut is mentioned in the fourteenth century

as a township (qasaba) in the Multan province: IM, 77, 93. For a brief allusion to Qubacha's defeat and

capture (clearly, therefore, not the campaign of 625/1228), see TN, I, 452, and BL ms., fol. 180a. According

to a doubtless anachronistic reference by Nasawi, 90 (tr. Buniiatov, 134), Sinan al-Din Chanisar, the Sumra

ruler of Daybul, was already subordinate to Iltutmish at the time of its seizure by Jalal al-Din's forces; for

his name, see Hodivala, Studies, I, 214-15.

56 TN, I, 446, and II, 4, 7, 9 (tr. 613, 723, 728, 731). For Aytemiir, cf. the spelling 'YTMR in JH,

I, 13; also p. 27, n.16 above. 57 For the attack on Qubacha, see Habibullah, Foundation, 95-6. The year 624 given at one point by Juzjani

for these events is an error for 625, as is clear from the context and from JH. see Hodivala, Studies, I, 205-

6. Taj (in ED, II, 242), which dates them in 624, is therefore wrong. Juzjani's month for the fall of Uchch

varies between Jumada I and Jumada

II, but I have adopted 27 Jumada 1/4 May 1228, since Iltutmish arrived outside the city on 1 Rabf i Iand the

siege lasted 2 months and 27 days: TN, I, 420, 446 (tr. 544, 612); JH, I, 10. 'Awfi, ibid., 10-22, gives a

more detailed account of the campaign. For Chanisar, see above, n.55.

~ legatees in the north-west. In 627/1229-30 the Delhi forces fell upon Ozbeg-bei, who fled to his master in

Iraq; whereupon Hasan Qarluq, among other local commanders, submitted to the sultan.58 The territory

controlled by these rulers is not known precisely. Numismatic evidence shows that Ozbeg-bei ruled for a

time in Binban, where he was succeeded by Qarluq. His principality must also have included Nandana,

Kujah (Gujrat), Sodra and Siyalkot, all lying in a tract where Jalal al-Din is known to have operated and

listed by Juzjani among Iltutmish's conquests.59 Qarluq is later found in control of Ghazna, Binban and

Kurraman. Nasawi describes the Delhi Sultan, on the eve of his attack on Ozbeg-bei, as master of the

territory 'up to the neighbourhood of the gates of Kashmir',60 suggesting that his frontier already stood on

the Jhelam. Iltutmish's last, abortive campaign, from which he returned a dying man in 633/1235-6, was

directed towards Binban, and may well have been intended to dislodge Qarluq, who had recently submitted

to the Mongols (see below, p. 105).61

While Iltutmish showed himself so attentive to his western frontier, he did not forget Bengal. Here the

bloodthirsty reign of cAli-yi Mardan, who had exhibited growing signs of insanity, ended i with his murder

in c. 609/ 1212-13. His successor, the more humane Husam al-Din cIwad, who assumed the style of Sultan

Ghiyath al-Din, lost Bihar to the Delhi forces.62 Some confusion exists regarding the history of the

Lakhnawti polity during these years, since from 616/1219 onwards we encounter a sultan named Mu'izz al-

Din 'Ali-yi 'Iwad, who was evidently Ghiyath al-Din's son and who was ruling either jointly with his father

or in succession to him.63 He is unfortunately not mentioned by Juzjani, who did nbt enter India, however,

until a few years later and hence is hardly a contemporary witness. According to that author, it was Ghiyath

al-Din who in 622/1225 fended off an invasion of Bengal by Iltutmish in person with an offer of tribute and

the recognition of his suzerainty. But shortly afterwards he reneged upon the agreement and once more

occupied Bihar. Iltutmish's eldest son and heir, 58 Nasawi, 92, 217 (tr. Buniiatov, 136, 267). For what follows, see Jackson, 'Jalal al-Din', 51.

59 TN, I, 452, to be corrected from BL ms., fol. 180a (KWJRAT), and IOL ms., fol. 243a (KJRAT): the two

mss. have respectively MWDWDH and MWDDH, clearly errors for SWDRH (cf. also Raverty's tr. 627). The

Mongol occupation of Nandana had, of course, been shortlived.

60 Nasawi, 217, ild mayalidarb Kashmir (tr. Buniiatov, 267).

61 TN, I, 449 (tr. 623); and cf. also I, 454-5 (tr. 631). On Hasan Qarluq's rule in Ghazna. Kurraman and

Binban, see ibid., II, 162 (tr. 1128-9); I. H. Siddiqui, 'The Qarlugh kingdom in north-western India during

the thirteenth century', IC 54 (1980), 75-91. 62 TN, I, 434-5, 437-8 (tr. 578-80, 590-1).

63 A Sanskrit epigraph of Vikrama samvat 1277, Jyastha ba. 15, Thursday (13 June 1219). from the Gaya

district of Bihar, refers to a Sultan 'Mojadina': ARIE (1962-3), 24, 80 (no. 261). For an inscription of 'AH

Shlr-i 'Iwad dated 618/1221, see Z. A. Desai, 'An early thirteenth century inscription from West Bengal',

EIAPS (1975), 6-12. That they refer to the same person is clear from coins of 620-1/1223-4 in the name of

Mu'izz al-Din 'AlI-yi 'Iwad: CCIM, II, 145-6 (nos. 3, 5). Nicholas W. Lowick, 'The Horseman type of

Bengal and the question of commemorative issues', JNSI35 (1973), 205-8, correctly argued for the name

'Ali here, though he wrongly assumed it represented 'AlI-yi Mardan.

~ Nasir al-Din Mahmud, who had at one time governed Lahore, as we have seen, and who now held the

iqta' of Awadh, thereupon led an army into Bengal. Taking advantage of 'Iwad's absence on a plundering

campaign in Kamrup (Assam), he was able in 624/1227 to seize Lakhnawti and then defeated and executed

'Iwad on his return. Iltutmish despatched a chatr to his son, who acted as the sultan's viceroy for less than

two years, dying in the first months of 626/the winter of 1228-9. Authority in the province was then

usurped by Ikhtiyar al-Din Dawlat Shah, also known as Bilge Malik, apparently a former officer of the

Delhi Sultan, until Iltutmish invaded Bengal and overthrew him in 628/1230-1.64 For the next twelve years,

at least, those regions of Bengal in Muslim hands were to remain subject to Delhi.

With the reoccupation of Lakhnawti, the Delhi Sultan became the only Muslim sovereign in India. His conquests enabled him to launch a new coinage, based on the pure silver tanga, which would in time

replace the dihliwals, imitations of the billon coins formerly minted by Hindu rulers at Delhi.65 Prior to this,

moreover, he had achieved an objective by which the majority of Muslim rulers set great store: recognition

from the 'Abbasid Caliph at Baghdad. It is now accepted that Iltutmish was the only Indian Muslim ruler

who received such recognition and that the titles borne by the Lakhnawti rulers were assumed unilaterally:

indeed, Iwad's assumption of the style Nasir Amir al-Mu'minirin ('Auxiliary of the Commander of the

Faithful') may have been one incentive for Iltutmish to take the title himself.66 Who initiated the

negotiations between Delhi and Baghdad, and at what point, is uncertain; but such information as we have

suggests that it may have been the caliph. We saw how al-Nasir endeavoured to incite the Ghurid Sultans

against his enemy the Khwarazmshah (p. 6); and two decades or so later Iltutmish, among others, might

have seemed similarly worth cultivating as a possible rival on the Khwarazmshah's southern flank.

For glimpses of the earliest diplomatic contacts with the 'king of India' we are indebted to Arabic

chroniclers writing in the west. The caliph sent out as his ambassador in 617/1220-1 the shaykh Radf al-

Din Abu'l-Fada'il al-Hasan b. Muhammad al-Saghani (d. 650/1252-3), who returned to Baghdad only in

624/1227, during the reign of al-Nasir's son al-Mustansir

64 On his coins he is called cAla' al-Dunya wa'l-Din Dawlat Shah b. Mawdud: CMSD, 21 (no. 53A). The

Berlin ms. of TN, fol. 99b,. confirms that Dawlat Shah and Bilge Malik were identical. Fullest reference to

the campaign of 628/1230-1 in JH, BL ms. Or. 2676, fol. 260. For events in Bengal, see Habibullah,

Foundation, 97-100, who was, however, unaware of the evidence cited in the previous note.

65 Simon Digby, 'The currency system', in T. Raychaudhuri and I. Habib (eds.), The Cambridge economic history of India, I (Cambridge, 1982), 95-6; Deyell, Living without silver, 179-80, 199-203,213-19.

66 A. H. Dani, 'Did Ghiyath-al-Din cIwad. Khalji of Bengal receive investiture from the Khalifah?', JPHS 3

(1955), 105-17; Roma Niyogi, 'A unique coin from Bengal and a review of cIwad's career', JNSI40 (1978-

9), 42-52.

~ (623-640/1226-1242). The new caliph sent him back to Delhi, and it is this latter mission which is

noticed by Juzjani, who in his brief survey of the 'Abbasids refers to the despatch of a mandate ('ahd) and a

banner (liwa') to Iltutmish in 625/1228, the very year of his own arrival at court, and says that he was

present at the celebratory banquet. Elsewhere he gives a fuller description of the embassy, which arrived in

Rabi I 626/February 1229 and brought Iltutmish robes of honour and a diploma confirming his authority

over all the territories he had conquered.67 Al-Mustansir also bestowed on the sultan - we must presume - the titles Yamin Khalifat Allah ('Right Hand , of God's Deputy') and Nasir Amir al-Mu'minin under which

he is exalted by a number of writers.68 For 'Awfi, he had become Khalifa-yi Amir al-Mu'minin ('Deputy of

the Commander of the Faithful').69 The usurper Iltutmish had thus attained respectability as one of the

family of orthodox Muslim princes whose, rule enjoyed the highest possible sanction. That he was no more

impervious to such honours than the Ghurids had been is clear from his assumption in 630/1233, following

the capture of Gwaliyor, of the quintuple nawbat, one of the attributes of full sovereign authority.70

Reasons for Iltutmish's triumph

Our survey of events has brought us to a point where the newly created Delhi Sultanate embraced a larger

territory than at any time prior to the last decade of the thirteenth century. There was nothing inevitable about this process. Much of it was the result of fortuitous circumstances over which Iltutmish had no

control. The Sultanate's survival was by no means guaranteed. Had it not been for the Mongols, Delhi, Sind

and the Panjab might have been swallowed up in the empire of the Khwarazmshah Muhammad. When he

destroyed that empire, Chinggis Khan inadvertently ensured that Muslim India would go its own way. In

establishing his own principality as the sole protagonist of Islam in the subcontinent, Iltutmish followed,

perhaps, a deliberate policy. Allowing his neighbours to weaken one another and intervening - as he did on

Qubacha's behalf against Jalal al-Din - only when it was absolutely necessary, he was then able to eliminate

these competitors one by one.

67 TN, I, 129 (section abridged in Raverty's tr.), 447, 454 (tr. 616, 629). Taj, in ED, II, 243. JH, BL ms. Or. 4392, fol. 128a. For the role of Saghani, who is not named in the Indian sources, see al-Safadl, al-Wafi bi'l-

Wafayat, ed. H.Ritter et al. (Damascus, 1931- ), XII, 241; al-Hawadith al-Jamia (early fourteenth century;

wrongly attributed to Ibn al-Fuwati), ed. Mustafa Jawad (Baghdad, 1351/1932), 262.

68 TN, I, 440, 450 (tr. 597, 624). Taj, fol. 217b. JH, I, 5. Jajarmi, preface to his tr. of Ghazali's Ihya" cUlumi'l-Din, BL ms. Or. 8194, fol. 3a .(with Mu'ln for Yarriin). Anonymous tr. of Razi's Sirr al-

Makhtuma, BN ms. Suppl. persan 384, fol. 2a.

69 JH, BL ms. Or. 2676, fols. 68a, 247b.

70 TN, I, 449 (tr. 620-1). Significantly, the Ghurid Sultan Ghiyath al-DIn Muhammad b. Sam had likewise

assumed the quintuple nawbat after the arrival of honorary robes from the caliph: ibid, I, 361 (tr. 383). For the nawbat, see Hodivala, Studies, I, 216-17.

~ This does not, of course, tell us what enabled Iltutmish to defeat his various rivals. The answer can only

be tentative, and must be based in large measure on developments on the Sultanate's western frontier,

where the information at our disposal, if not plentiful, is at least fuller than that pertaining to Bengal. In

612/1215, if we are to believe Ibn al-Athir, Qubacha had encountered Yildiz's invading army with 15,000

men; against Jalal al-Din seven years later he mustered a force that was certainly no greater than 20,000

and may well have been half that size.71 These numbers stand in sharp contrast with the (doubtless

exaggerated) figure of 130,000 for the army which Iltutmish is said to have raised for his own campaign

against the Khwarazmians.72 Such statistics are notoriously hazardous guides to the size of medieval

armies, and perhaps they are in the last analysis unusable. Yet we cannot discount the possibility that the Delhi ruler presided over a significantly larger military establishment than did his neighbour and rival, who

had been obliged to disburse considerable quantities of treasure during the Mongol siege of Multan.73 It is

hardly coincidental that Iltutmish embarked on the reduction of Qubacha in the immediate wake of three

highly successful campaigns: that of 622/1225 against 'Iwad in Bengal, yielding the substantial sum of

eighty laks (8,000,000) of silver (presumably dirhams) in tribute,74 and the expeditions of 623-4/1226-7

which resulted in the capture of the Hindu fortresses of Ranthanbor and Mandor. The specie obtained by

such victories would surely have enhanced Iltutmish's capacity to recruit more formidable armies.

Of the fact that a ready pool of military support lay to hand for the ruler whose resources enabled him to

pay for it, there can be little doubt. We have seen how prior to this, in the era of Mu'izz al-Din, soldiers of

fortune such as Muhammad b. Bakhtiyar hired themselves out to the highest bidder and launched initiatives

of their own. Although Fakhr-i Mudabbir undoubtedly exaggerates the enhancement in their material prosperity awaiting those who migrated to Muslim India,75 it is likely that the region was coming to be seen

as some sort of El Dorado. The irruption of the Mongols into the eastern Islamic lands after 617/1220 must

have considerably increased the number of adventurers - both Turks and 'Tajiks' - eager for whatever

enterprise was on offer. We may imagine a veritable reservoir of unattached warriors and officials in north-

western India in the 1220s. Many of 'the chief men of Khurasan, Ghur and Ghazna' secured a refuge in the

first instance at the court of Qubacha, according to Juzjani, who himself arrived at Uchch in 624/1227 and

was made qadi-yi lashgar to Qubacha's

71 IA, XII, 203/311, for the battle with Yildiz. Nasawi, 88 (tr. Buniiatov, 133), for the Khwarazmians, with

10,000; TJG, II, 146 (tr. Boyle, 414), has 20,000, but this perhaps includes the reinforcements sent by

Iltutmish (above, p. 33). 72 Nasawi, 90 (tr. Buniiatov, 134): 30,000 horse and 100,000 foot.

73

TN, I, 420 (tr. 537-8). 74

Ibid., I, 438, 445 (tr. 593, 610). 75

SA, 20.

~ son.76 In the previous year the maliks of Ghur had finally abandoned their homeland to the Mongols and

fled to join Qubacha.77 But others among these distinguished condottieri made for Delhi: they included one

of the few surviving members of the Ghurid dynasty, Nasir al-Din Abu Bakr b. Suri, who became one of

Iltutmish's maliks and died at Delhi in 620/1223.78

There were also not a few whose loyalties shifted. It was a time for highly volatile allegiances. In the

course of Jalal al-Din's negotiations with Iltutmish, two of his envoys, weary of the hardships they had been

required to undergo, deserted their master and entered the service of the Delhi Sultan.79 Qubacha in

particular seems often to have been the victim of such transfers. Early in the confrontation with Jalal al-

Din, two important maliks - Nusrat al-Din Muhammad b. Husayn b. Kharmil and Taj al-DIn Yinaltegin

(the future ruler of Sistan) - left him for the Khwarazmians.80 What became of the maliks of Ghur who

flocked to Qubacha's court in 623/ 1226, we are not told; but Qutb al-Din Hasan b. 'A1i, who may have

been one of their number,81 is subsequently found at Iltutmish's court. Even Qubacha's waztf, 'Ayn al-Mulk

Husayn al-Ash'ari, was within a short time appointed to a similar position at'the court of Iltutmish's son

Rukn al-Din Firuz, then muqta' of Bada'un.82

It may well be that in time the Delhi Sultan appeared to offer better prospects to adventurers, whether military men or those of a more scholarly persuasion, than did his neighbours. Not the least attractive

features of the Sultanate to immigrants would have been its geographical location, making possible

lucrative raids into Hindu territory, like those on Ranthanbor and Mandor, which provided booty for the

mercenary, and its need of experienced officials, which afforded employment for the savant. Juzjani

doubtless made some such calculation when he quitted Uchch in Safar 625/January 1228 to join the Delhi

forces under Kezlik Khan outside the walls three months before the citadel surrendered, and took care to

wait upon Iltutmish on the very day of his arrival with the

76 TN, I, 420 (tr. 541-2), with the month Jumada I/May for his arrival, whereas at I, 446 (tr. 611-12), he

gives Rajab/June-July; for his emigration from Ghur, see also II, 184-5 (tr. 1203-4). On the distinguished

immigrants from Khurasan, see I, 419 (tr. 534). 77 Ibid., I, 420 (tr. 539-41): a year and a half after the Mongol investment of Multan, i.e. early in 623/1226.

78 Ibid., I, 340 (tr. 345); he is listed among Iltutmish's maliks at I, 451 (tr. 626). He belonged to the branch

that ruled the petty principality of Madin.

79 Nasawi, 91 (tr. Buniiatov, 135). ,

80 Nusrat al-DIn: ibid., 88, 140, 141 (tr. Buniiatov, 132, 186, 188). Yinaltegin: TN, I, 284 (tr. 200); and for

his subsequent career, C. E. Bosworth, The history of the Saffarids of Sistan and the maliks ofNimruz

(2471861 to 949/1542-3) (Costa Mesa, California, 1994), 407-10.

81 TN, II, 135, 140-1 (tr. 1061, 1070-1): Qutb al-Din set out for 'Hindustan' in 620/1223 along with other

maliks, though his companions were allegedly all slain by the Mongols en route. For the 'Husayn' of the

printed text, BL ms. reads 'Hasan' throughout (fols. 183a, 185b, 193a, 220a, 240).

82 TN, 1,454 (tr. 631).

~main army soon afterwards.83 Awfi too seems to have been ready enough to abandon his benefactor

Qubacha. A scholar who himself reached Delhi in 620/1223 speaks of the flight from the Mongols to

Iltutmish's court of 'Muslims of Khurasan, of Transoxiana (Ma wara' al-Nahr), of Ghur, of Ghazna, nay all

the Muslims of the east (bed kaffa-yi musulmdndn-i mashriq)'.84 In this context, Juzjanl's encomium on the

sultan - that he was ever generous in his gifts to landowners (dahaqin) and strangers from great cities (ghuraba-yi amsdr), and that he made his capital a haven for those escaping from the Mongol deluge85 - is

not without significance. And his verdict is echoed by authors of the following century, notably 'Isami, who

speaks of the arrival of 'sayyids from the Arab lands and 'ulama' from Bukhara'.86 Iltutmish is the only

Delhi monarch, other than Muhammad b. Tughluq in the fourteenth century, who is known to have sought

deliberately to attract immigrant notables in this fashion. It is an intriguing possibility that he benefited

from widespread desertions by many of those who had only recently obtained asylum in Sind and who

judged that Qubacha was not the power most likely to guarantee their future security or, indeed, prosperity.

The immigrant nobility under Iltutmish At the end of his chapter on Iltutmish, Juzjani furnishes a list of his nobles, including Ghuris, Turks and

'Tajiks'. The great majority are merely names to us; they are mentioned neither elsewhere in the Tabaqdt-i

Ndsiri nor in any other source. Some, however, are met with in the body of the chapter, and others, who

were the sultan's own ghulams, are also accorded biogra- phies in tabaqa 22: to these we shall return

shortly. A few correspond to amirs specified by Hasan-i Nizami as having supported Iltutmish in the fight

against Aram Shah's party or having not long afterwards accompanied him on his Jalor campaign: 'Izz al-

Din Bakhtiyar, commemorated in one of the very earliest epitaphs so far discovered in the Sultanate's

territories; Nasir al-Din Mardan Shah; and Iftikhar al-Din Muhammad-i 'Umar, described by Juzjani as the

chief amir (malik al-umara) and commander at Kara.87 We can even detect a certain continuity with the era

of the Ghurid conquests. Rukn al-DIn Hamza, named by both authors as one of

83 Ibid., I, 447, and II, 3 (tr. 615, 722-3). For Kezlik Khan's title (Tu. kezlik, 'small knife'), see Clauson, Etymological dictionary, 760; TMENP, IV, 3-4 (no. 1714).

84 Jajarmi, BL ms. Or. 8194, fol. 3b. See also the Persian translation of Birum's al-Saydanafi'l-Tibb, BL ms.

Or. 5849, fol. 4a; Nazir Ahmad, 'Beruni's Kitab-as-Saydana and its Persian translation', Indo-Iranica 14,

part 3 (1961), 17.

85 TN, I, 440-1 (tr. 598-9). 86 FS, 114-15 (tr. 226-7). TFS, 27.

7 Taj, fols. 221b, 229a. Juzjani's list of maliks is at TN, I, 452 (reading KWH for KRH, but cf. Raverty's tr.,

627). For 'Izz al-DIn Bakhtiyar (d. 616/1219), see Iqtidar Husain Siddiqi, 'Historical information in the

thirteenth century collections of Persian poems', Studies in Islam 19 (1982), 57-8.

~ Qtutmish's amirs, is possibly identical with the homonymous figure active in Mu'izz al-Din's service and

employed on an embassy to Prthvlraja III in 587/1191.88 It is more certain that both 'Izz al-Din 'A1I, at one

time muqta' of Nagawr, and Husam al-Din Oghulbeg in Awadh had held office since Mu'izz al-Din's

reign.89

What became of most of the old nobility is obscure. As we have seen, Qutbi amirs had supported Iltutmish

at the outset, and Juzjani's list includes two amirs with the sobriquet 'Qutbi', presumably former ghulams of

Aybeg. But the Mu'izzi amirs had rallied to Aram Shah, which is why the Mu'izzis mentioned by Fakhr-i

Mudabbir as among Aybeg's entourage at Lahore in 602/1206 are lost to sight thereafter.90 It is accordingly

possible that Iltutmish was obliged to constitute in effect a new class of high-ranking officers. Immigrant

notables would in time have furnished him with the means of doing so. Some were Turkish grandees of free status. Hasan-i Nizami indicates that in meeting *Berki's attack Iltutmish benefited from the support of

Sayf al-DIn FIruz, who is to be identified with a cousin of Yinaltegin called by Juzjam Firuz-i Iltutmish and

FIruz b. Salar, and who was apparently a warlord originating from the Qangli confederacy in the steppes

north of Khwarazm.91 cAla' al-Din J-ani, bombastically described by Juzjani as a 'prince of Turkistan',

received La'khnawti as'his iqta' following Iltutmish's victorious campaign of 628/1230-1 and is

subsequently found at Lahore.92 The Ghurl malik Qutb al-DIn Hasan became comptroller of Iltutmish's

household (wakil-i dar); and 'Izz al-Din Muhammad Salarl, who was also probably a Ghuri, served as

barbeg (i.e. amir-hajib).93 We have seen how bureaucrats as well as soldiers sought asylum in India from

the Mongol onslaught, and among those in office at Iltutmish's death are men with cognomina from

Khurasan like Shafurqani and Tayaqani;94 though whether they were first-generation immigrants is

unknown. A similar 88 Rukn al-DIn does not appear in the list of maliks as given in Hablbl's text of 77V, but cf. IOL ms., fol.

242b, and Raverty's tr., 626. Taj, fols. 77b-78a, 229a; for the embassy to Prthvlraja, see fol. 38a.

89 For 'A1I and Oghulbeg in Mu'izz al-Din's reign, see TN, I, 422-3 (tr. 549), and pp. 26-7 above. For 'A1I

among Iltutmish's maliks, TN, I, 452; and for Oghulbeg, ibid., I, 451 note, and cf. BL ms., fol. 179a (tr.

627); also JH, BL ms. Or. 2676, fols. 263b-264a (Nizamu'd-din, Introduction, no. 1729). By 620/1223

Nagawr was held by the otherwise unknown Karim al-DIn Hamza: TN, I, 284 (tr. 200; but BL ms., fol.

179b, reads Najlb al-Din).

90 SA, 25, 73. They are the sipahsalars Husam al-Dawla wa'l-Din Ahmad (-i ?) 'AH Shah, Mubariz al-

Dawla wa'l-DIn Toghriltegin 'A1I (-yi ?) Hasan, and Asad al-Dawla wa'l-Din 'A1I (-yi ?) Muhammad

Abu'l-Hasan. The last two are both entitled ulugh ('great') dadbeg (i.e. amir-i dad).

91 Taj, fol. 221b. For Firuz, see also TN, I, 284, 299 (text reads 'Nimrtiz' in error), 452 (tr. 199, 235, 625).

His father's sister was the wife of the Khwarazmshah Il-Arslan and hence grandmother to the Ghurid

Sultans' enemy 'Ala' al-DIn Muhammad? Nasawi, 36 (tr. Buniiatov, 81), confirms the relationship. TN, I,

298 (tr. 235), gives a garbled account of this family.

92 Ibid., I, 448,452, 455, and II, 9 (tr. 618, 626, 634, 731-2).

93 TFS, 39. For Salari, see TN, I, 446 (tr. 613), and JH, I, 12; also Habib, 'Formation', 13. 94 TN, I, 456, and II, 30 (with SRQANY in error; cf. tr. 635, 761).

~ obscurity shrouds the antecedents of Iltutmish's wazir, Mu'ayyad al-Mulk (later Nizam al-Mulk) Qiwam

al-Dawla wa'l-Din Muhammad b. Fakhr al-Mulk Sharaf Abi Sa'd Junaydi, and the sultan's 'arid, 'Imad al-

Mulk Sharaf al-Dawla wa'l-DIn Abu Bakr, of whom we learn only that he was of illustrious lineage and

was not a Turk.95 But whatever their origins or talents, no Muslim ruler would have felt easy in relying

exclusively on adherents of free status. Like Mu'izz al-Din, Iltutmish took care to build up a corps of

Turkish slaves (Persian bandagdn; sing, banda), known as the 'Shamsis', whose loyalty was focused on him

alone. Under his successors, they would come to play a more prominent role in the government of the

Sultanate. 95 For the wazir's full name and style, see JH, BL ms. Or. 4392, fol. 128a (cf. also ms. Or. 2676, fol. 68a);

Jajarmi, BL ms. Or. 8194, fol. 3a. Barani later heard that Junaydi was of plebeian origin, the grandson of a

weaver (julaha): TFS, 39. 'Imad al-Mulk: JH, BL ms. Or. 2676, fols. 263b-264a.

~CHAPTER 3

Sultans and sources

Shams al-Din Iltutmish died on 20 Sha'ban 633/29 April 1236. In contrast with Aybeg, he founded a

dynasty, which ruled until Iltutmish's own slave Balaban,1 hitherto viceroy (na'ib) to Nasir al-DIn Mahmud

Shah b. Iltutmish, usurped the throne in 664/1266 and reigned as Sultan Ghiyath al-Din. Balaban's dynasty too was shortlived. After the brief reigns of his grandson Mu'izz al-Din Kayqubad and the latter's infant son

Shams al-DIn Kayumarth, its life was snuffed out in 689/1290 by the Khalaj officer Jalal al-Din Firuz, the

first of the Khalji Sultans. Designations like 'Slave kings' and 'Slave dynasty', traditionally applied to the

thirteenth-century Delhi Sultans, are misnomers. Only Iltutmish and Balaban were ghulams; the majority of

the rulers, their respective descendants, had at no time been slaves. In this book, therefore, the two

dynasties will be termed 'Shamsids' and 'Ghiyathids' in the interests of greater accuracy.

Juzjani and the Shamsids

The historian of the thirteenth-century Delhi Sultanate is not embarrassed by a wealth of literary sources.

With the exception of the accounts of the Khwarazmian and Mongol operations in India given by Juwayni and Nasawi, no external source has survived from the period between Ibn al-Athir and the end of the

thirteenth century which refers to contemporary events in the subcontinent. One reason may well be the

lack of contact with the Caliphate. That the reigns of Iltutmish's first two successors were noticed in the lost

work of the Baghdad historian Ibn al-Sa'i (d. 674/1276), we learn from a citation by a mid-fourteenth-

century chronicler writing in Mamluk Syria.2 Ibn al-Sa'I had presumably derived his information from the

caliphal envoy Saghani, who left India for Baghdad in 637/1239 after a

1 For balaban, 'sparrow-hawk', see Sauvaget, 'Noms et surnoms', no. 61.

2 (al-Mufaddal) Ibn Abi'l-Fada'il (c. 1350), al-Nahj al-Sadid, partial edn and tr. Samira Kortantamer, Agypten und Syrien zwischen 1317 und 1341, IU 23 (Freiburg i. Br., 1973), Ar. text 28-9 (German tr. 107).

There is a virtually identical passage in the fourteenth-century Baghdad chronicle al-Hawddith al-Jdmica,

104. On Ibn al-Sa'i, see above, p. 18, n.56.

~ stay of eleven years.3 Whether the last 'Abbasid Caliphs exchanged embassies with Iltutmish's successors,

we are not told. The Delhi monarchs continued to employ the style 'Auxiliary of the Commander of the

Faithful' (Nasir Amir al-Mu'miriin) on their coins down to the extinction of the Baghdad Caliphate by the

Mongols in 656/1258 and even beyond.4 But it is unlikely that this rested on official conferment: had his

patron Nasir al-Din Mahmud Shah, at least, received confirmation of his title from Baghdad, Juzjani would

assuredly have told us. It is a measure of the Sultanate's isolation during this period that chroniclers in

Mamltik Egypt, who periodically listed contemporary foreign rulers at the head of an annal, named the

Delhi Sultan correctly only once (in 662/1264) in the course of the period from 635/1237 onwards.5 Not until 700/1300-1, after the invasion of Gujarat by the forces of 'Ala' al-Din Khalji had opened up a new

channel of communication with the Mamluk empire, did its chroniclers again include the sultan's name at

regular intervals in lists of contemporary sovereigns.6

Although a number of references to major figures and historical events can be gleaned from the works of

Persian poets such as Siraji and 'Amid Sunnami, who graced the sultan's court or those of his Muslim

neighbours in India around the middle of the century,7 for the era of Iltutmish's progeny down to 658/1260

we are overwhelmingly dependent on a single narrative source, the final three sections of Juzjani's

Tabaqat-i Nasiri. Of all the historians of the Sultanate, Juzjani had the best vantage-point from which to

observe events, since he occupied on three occasions one of the highest civil offices, that of grand qadi

(qadl al-quddat) of the empire. Yet though comparatively rich in data for this period, the Tabaqat is not an easy work to use. Some events are recounted in such opaque terms that their significance is almost

completely lost. The arrangement of the material is also extraordinarily confused: the same episode may be

described twice at different points - both under the relevant reign in tabaqa 21 and in one or more of the

biographies of Shamsi slaves that make up tabaqa 22 - but with varying and indeed conflicting details. It is

as if the chronicler's aim was to camouflage rather than to illuminate events. This is all the more regrettable

given the absence of any alternative sources. Juzjahi's work was quarried by

3 al-Safadl, wafi; XII, 241.

4 The caliph's name was first omitted on the coins of the Khalji Sultan Rukn al-Din Ibrahim (695/1296):

Edward Thomas, Chronicles of the Pathan kings of Delhi (Delhi, 1871), 255.

5 Ibn al-Dawadari, VIII, 102. Ibn Abi'l-Fada'il, al-Nahj al-Sadid, partial edn and tr. E. Blochet, 'Moufazzal

Ibn Abil-Fazail. Histoire des Sultans Mamlouks', Patrologia Orien-talis 12-20(1919-29), 123.

6 Al-Yunini (d:"726/1326), al-Dhayl 'ala Mir'ati'l-Zaman, TSM ms. Ill Ahmet 2907/e.3, fols. 196a (700;

with incorrect name), 210b (701); III Ahmet 2907/e.4, fols. 25b, 36a, 43b, 157b, 179a, 212b (with varying

degrees of inaccuracy). Ibn Abi'l-Fada'il, ed. Blochet, 534-5, 556, also begins naming the Delhi Sultan

(incorrectly) in 700.

7 Siddiqi, 'Historical information'. TFS, 113, mentions Khwaja Shams-i Mu'in, who wrote 'volumes'

(mujalladat) in praise of Qutb al-DIn Hasan Ghuri, but his work has not survived.

~ 11 the later authors who cover these years, and it is only rarely that one of them - 'Isami, for instance, or

the fifteenth-century chronicler Sirhindi -upplies any additional information, its provenance and reliability

alike far rom certain.

This tendency towards obfuscation is illustrated by Juzjam's treatment of he succession to Iltutmish. The

sultan's eldest son, Nasir al-DIn Mahmud shah, who died in Lakhnawti in 626/1229, had been widely

expected to succeed him. Following the Gwaliyor expedition in 630/1233, Rukn al-Din Firuz Shah, as the

next son, had been appointed muqta' of Lahore, a position once occupied by Nasir al-DIn Mahmud. FIruz

Shah accompanied [ltutmish back to Delhi not long before the sultan's death, as if he was being groomed

for the throne, and Juzjani confirms that the eyes of the people were on the prince.8 A work composed in Iltutmish's last years appears to corroborate this, since it is dedicated to the sultan and FIruz Shah jointly, as

if the latter were heir-apparent.9 Firuz Shah duly ascended the throne within a few days of his-father's

death. But JuzjanI at one point alleges that in the wake of the Gwaliyor expedition Iltutmish had marked

out for the succession his eldest daughter Radiyya, who may thus have been his firstborn child and whose

mother was his chief wife, and had caused a diploma to be drawn up in her favour. When certain officials

objected, he allegedly predicted that none of his sons would be found fit to rule.10 It is noteworthy,

however, that JuzjanI was at Gwaliyor at this time and did not return to Delhi until 635/1238 (i.e. during

Radiyya's reign):11 he could not have been present, and he does not in fact claim to have seen the diploma.

In these circumstances, the story may well be apocryphal and have been circulated by those who enthroned

her: according to the fourteenth-century Moroccan visitor Ibn Battuta, her tomb had become an object of

pilgrimage,12 and it is noteworthy that, as we shall see (pp. 69-70), Balaban himself was indebted to her for

his first promotion to office. Subsequently, in view of its disparagement of Iltutmish's sons (including, of course, the reigning sultan Nasir al-Din Mahmud Shah),13 the tale could have acquired a new significance

at the time JuzjanI was writing, when the displacement of the Shamsid dynasty was perhaps already on the

horizon. .

Juzjani's own views of Iltutmish's first four successors can be gleaned from the Tabaqat. Firuz Shah (633-

634/1236) was a pleasure-loving youth who left the reins of government to his mother, the energetic and

vindictive Shah Terken. His brief reign was dominated by a revolt on the part of a

8 TN, I, 454-5 (tr. 630, 631).

9 Anonymous, tr. of Razi's Sirr al-Makhtuma, BN ms. Suppl. persan 384, fol. 2a. 10 TN, I, 458 (tr. 638-9). Nizami (in HN, 230-1) believes that Iltutmish originally designated Radiyya, but

then changed his mind and groomed FIruz Shah instead.

11 TN, I, 448-9, 460 (tr. 620, 643-4). 12 IB, III, 169 (tr. Gibb, 632).

13 As Nizami points out, in HN, 230-1 n.84; cf. also his On history and historians, 84. It is unfortunate that

elsewhere (HN, 253, 256) Nizami's insight is impaired by his acceptance ot cIsami's testimony that

Mahmud Shah was not Iltutmish's son but his grandson.

~group among Iltutmish's senior amirs, including the wazir Junaydi, who may have supported FIruz Shah's

brother Ghiyath al-DIn Muhammad as a candidate for the throne.14 The rebellion was eventually put down by Radiyya, who had been enthroned in his place. We are clearly intended to draw a contrast both with

Firuz Shah's elder brother, the 'wise and prudent' (farzana-u 'aqil-u bikhrad) Nasir al-DIn Mahmud Shah,

whose premature death can be seen as a heavy blow to the Sultanate,15 and with his successor. Radiyya

(634-7/1236-1240) is credited with all the attributes of a successful ruler except one, namely that she was

not a man; she is the only Shamsid sultan whom Juzjani describes as a war leader (lashgarkash).16

Tradition makes much of Radiyya's adoption of masculine garb and her public appearances riding on an

elephant.17 But whatever the 'ulama' thought of this, it is clear that Radiyya's backers had intended her to be

a figurehead and that her offence lay in her growing self-assertiveness. Initially her coinage, on which her

father's name was associated with her own, had testified to her insecurity; but in c. 635/1237-8 Iltutmish's

name was dropped.18 Deposed in favour of her brother Bahrain Shah, she was imprisoned in Tabarhindh

and killed in 638/1240 in a vain bid to recover the throne.

Of Mu'izz al-DIn Bahrain Shah (637-9/1240-2) we are told that he was a courageous sovereign who had a

penchant, however, for shedding blood.19 He was overthrown when many of his commanders mutinied and

stormed Delhi. Bahrain Shah had become highly unpopular with the 'ulama', one of whose number he had

executed and who participated in an abortive conspiracy to dethrone him; the shaykh al-islam, whom he

sent out to negotiate with the rebel amirs, went over to the enemy.20 In these circum- stances, Juzjani's

partiality for the sultan is difficult to understand; but the reason may be nothing more complex than that he

owed to Bahrain Shah his first appointment as grand qadi.21 The events culminating in Bahram Shah's

overthrow appear to have sickened him, since he resigned his office and left Delhi for Lakhnawti, where he

remained for over two years. He was thus absent during the early years of FIruz Shah's son 'Ala' al-Din Mas'ud Shah (639-44/1242-6), who; 'we are blandly assured, was generous, right- thinking and endowed

with every laudable quality. But the new sultan in turn1 fell under evil influences and took to executing his

maliks and amirs,22 so that his uncle, Iltutmish's youngest son Nasir al-Din Mahmud Shah, was secretly

invited to supplant him.

Of the various changes of sovereign noticed in the Tabaqat, this particular

14 Siddiqi, 'Historical information', 56-7. 15 TN, I 453, 454 (tr. 628, 630).

16 Ibid., I, 457 (tr. 637-8). 17 Ibid., I, 460 (tr. 643). IB, III, 167 (tr. Gibb, 631).

18 CMSD, 40 (nos 161, 161 A): apparently coins struck in Lakhnawti, however, bore Radiyya's name alone throughout, ibid., 41 (nos 161B-D).

19 TN, I, 462 (tr. 649). 20 Ibid., I, 464, 466, 467 (tr. 652, 657, 658-9).

21 Ibid., I, 466 (tr. 657-8). 22 Ibid., I, 468, 471 (tr. 660, 668-9).

~ oup is the most obscure - not surprisingly, perhaps, since the monarch who now ascended the throne is

the one to whom Juzjani dedicates his listory. Mahmud Shah (644-664/1246-1266) is the most shadowy of

all the hirteenth-century sultans.23 The era appears to be dominated by Juzjanl's )atron, Baha' al-Din

Balaban-i Khwurd ('the Lesser'), entitled Ulugh Khan, the future sultan. Balaban, whose daughter the sultan

married and who acted as viceroy (na'ib), with a brief interval of about one year, from 547/1249 until Mahmud Shah's death, seems to play Earl Godwin to Mahmud Shah's Edward the Confessor. Juzjani says

that the sultan possessed 'the qualities of saints and the characteristics of prophets' (awsaf-i awliya wa-

akhlaq-i anbiya), and includes among his many virtues piety, faith, asceticism and continence (taqwa-u

diyanat-u zahadat-u siyanat).24 Professor Nizami has suggested that Jtizjani constructed this picture in

order to justify Balaban's dominance.25 Mahmud Shah's image was undoubtedly persistent. Tales circulated

in the following century that he had found an outlet for both his energies and his piety in calligraphy: he

copied Qur'ans and purchased his food with the proceeds.26 His austere lifestyle even attracts comment

from a later chronicler writing outside India.27 It is accordingly difficult to avoid the impression of a

monarch who was somewhat detached from his own-court. Factions jostle for power at the centre and for

the most desirable iqta's; leading grandees are ruthlessly cut down or sent into exile in the provinces;

behind a facade of military expeditions and conspiratorial intrigues the figure of Balaban, as the sultan's

deputy, is never far away. But what Nasir al-Din Mahmud Shah thought of all this -whether, for instance, he welcomed or deplored either Balaban's fall from power in 651/1253 or his reinstatement in the following

year - we cannot discern.

Juizjani wrote'only a few years prior to Balaban's accession. It is Balaban who receives by far the

longest biography of the twenty-five Shamsi ghulams, and Juzjanl's expressions of gratitude to him for gifts

and pensions recur frequently1 in the book.28 Even if we accept the tradition attributed to Balaban, and

found in the hagiographical Sarulr al Sudur, that Juzjani the qadi did not fear him,29 Juzjani the chronicler

seems, nevertheless, to have felt inhibited from revealing circumstances which cast his benefactor in a poor

light. Nor is he able or willing to do full justice to Balaban's enemies: although Kushlu Khan ('Izz al-Din

Balaban) is the subject of a biography in tabaqa 22 and is praised for his favour towards the 'ulama' and

ascetics,30 23 For a survey of the reign, see Mohibbul Hasan, 'Mahmud I, Nasir al-DIn', Enc.Isl.2.

24

TN, I, 477 (tr. 674). 25

Nizami, On history and historians, 82-3.

26 TFS, 26. FS, 156 (tr. 280-2). IB, III, 169 (tr. Gibb, 632). DR, 50, describes Mahmud Shah as 'immersed

in the affairs of God'.

27

Ibn Abi'l-Fada'il, ed. Kortantamer, Ar. text 29 (German tr. 107).

28 TN, I, 481, and II, 2, 52-3, 61, 62, 69 (tr. 681, 720, 808, 821-2, 823, 835).

29 Cited in Nizami, On history and historians, 93. 30 TN, II, 36 (tr. 775-6).

~ the na'ib's other great rival, Qutlugh Khan, is accorded no such distinction and is referred to only in

passing. At this time, moreover, the proximity of the pagan Mongols both threatened the integrity of the

Sultanate and afforded an incentive to refractory grandees to defy the Delhi government. Balaban was

restored to power in 652/1254 as a result of manoeuvres in which certain of his confederates were in league

with the Mongols. But Juzjani's own account is noticeably coy on the subject, and were it not for the details

on India furnished by authors, like Wassaf and Rashid al-Din, writing in the dominions of the Mongol

Ilkhan in Persia, we should have little idea of the complexity of these events.

Sources after 658/1260

After 658/1260 Juzjani's voice falls silent, and we enter upon an era for which genuinely primary source

material is extremely meagre. To write a connected account of the reigns of Balaban (664-85/1266-87), of

his grandson Mu'izz al-Din Kayqubad (685-9/1287-90) or of Jalal al-Din Firuz Shah Khalji (689-95/1290-

6) is even more difficult than for the fourteenth-century sultans, since there are no contemporary narrative

sources.31 In large measure we are dependent either on authors who cover a considerable period but who

wrote in the middle of the fourteenth century, or on those who composed shorter works to commemorate

specific events. The exception is a historical tradition that was current in Persia by the end of the century.

This, the earliest survey of the period down to 'Ala' al-Din Khalji's reign which has survived, is found in the

brief history of the Sultanate which Wassaf inserted in his Tajziyat al-Amsar (designed as a sequel to Juwayni s work) in or just before 702/1303 and presented to the Ilkhan Ghazan. This part of Wassaf's work

was copied within the next year or so into the Indian section of the great historical encyclopaedia, Jami' al-

Tawarikh, of the Ilkhanid wazir Rashid al-DIn Fadl-Allah al-Hamadani.32 Rashid al-Din added the odd

detail of his own; though his statement that Uchch and Multan are governed by the sultan's son indicates

that some of his information dated from the reign of Jalal al-Din Khalji or perhaps even that of Balaban.33

It was under Balaban that the celebrated poet Yamin al-Din Abu'l-Hasan, better known as Amir

Khusraw Dihlawi (b. 651/1253; d. 725/1325),

31 For these reigns, see HN, 277-325; Habibullah, Foundation, chaps. 7-8; K. S. Lai, History of the Khaljis

AD. 1290-1320, 3rd edn (Delhi, 1980), chap. 2.

32 For the date at which this section of his work was presented to the Ilkhan, see Wassaf, Tajziyat al-Amsar

wa-Tazjiyat al-A'sar, lithograph edn (Bombay, 1269/1853), 405. Rashid al-Din's Indian chapters were

composed over the years 702-3/1302-4: see JT, ed. Karl Jahn, Die Indiengeschichte des Rasid ad-Din

(Vienna, 1980), introduction, 9.

33 Ibid., Pers. text Taf. 13, Ar. text Taf. 51 (German tr. 36). Rashid al-DIn also states that Bengal is under

the rule of a cousin of the Delhi Sultan who has repudiated his authority: Pers. text Taf. 15-16, Ar. text Taf.

52 (tr. 39).

~ Who was the son of one of Iltutmish's ghulam troopers, began work on his first diwans, Tuhfat al-Sighar

(c. 671/1272-3) and Wasat al-Hayat (although his was not completed perhaps until c. 690/1291). These and his third Iwdn, Ghurrat al-Kamal (693/1294) occasionally allude to contemporary vents, and the preface

(dibacha) to the last-named work contains a valuable utobiographical sketch. During the reigns of

Kayqubad and Jalal al-Din Chalji, Khusraw composed his earliest epic narrative poems (mathnawis),

espectively Qiran al-Sa'dayn, centred on the reconciliation between Kay- qubad and his father Bughra

Khan in 686/1287, and Miftah al-Futuh, commemorating the victories of Jalal al-DIn in 690/1291.

Khusraw's principal defect - excessive adulation of the reigning sultan - is amply illustrated in the opening

of his one prose work, the Ta'rikh-i 'Ala'J or Khaza'in al-Futuh (711/1311-12), where a bland account of

the accession of Jalal al-Din's nephew 'Ala' al-Din Khalji in 695/1296 omits all mention of the old sultan's

murder. In the short but occasionally useful sketch of the Sultanate's history from Iltutmish onwards, with

which he prefaces his Diwal Rani or 'Ashiqa (centred on the love between 'Ala' al-Din's son Khidr Khan and a Hindu princess), he could afford to be more forthright, since by the time he completed the poem, in

720/1320, 'Ala' al-Din and his sons were all dead.34

These works apart, we are thrown back on the Ta'rikh-i Firuz-Shahi of Diya'-yi Barani35

(completed in 758/1357) and the epic Futuh al-Salatin of Isami (750/1349), together with the baldly

annalistic Ta'nkh-i Mubarak-Shdhi of Yahya b. Ahmad Sirhindl, who wrote as late as 838/1434. The

sources available to 'Isami and Sirhindl are unknown; the latter may possibly have used the now lost

continuation (mulhaqat) of the Tabaqat-i Nasiri, attributed by the seventeenth-century compilator Firishta

to 'Ayn al-Din Bljapuri (d. 795/1393), who like 'Isami was a subject of the breakaway Deccan Sultanate.36

About Barani's sources we are better informed. He claims for the period prior to Jalal al-DIn Khalji's

accession that he is relying on hearsay from his father and uncle, who were officers in the service of the

first two Khalji sovereigns, and his maternal grandfather, Husam al-Din, who had been comptroller of the household (wakil-i dar) of Balaban's barbeg (amir-hajib) and whom that sultan subsequently appointed as

governor (shihna) of Lakhnawti.37 Sometimes he attributes his informa-

34 On these works, see M. Wahid Mirza, The life and works of Amir Khusrau (Calcutta, 1935).

35 He is called Diya" al-Din Barani by later authors: Irfan Habib, 'Baranl's theory of the history of the Delhi

Sultanate', IHR 1 (1980-1), 99 n.l; Muhammad Bihamadkhanl. Ta'rikh-i Muhammadi (fifteenth century),

BL ms. Or. 137, fol. 409b.

36 Firishta, Gulshan-i Ibrahimi, lithograph edn (Bombay, 1247/1831-2, 2 vols.), I, 5, 131, 165. On Bljapuri,

see A. T. M. cAbd al-Jabbar, Mahbub dhi'l-Manan Tadhkhira Awliya Dakkan (Hyderabad, Deccan, 1332/1914), 538-41, who claimed to have possessed a ms. of the Mulhaqdt which was subsequently lost.

37 Husam al-DIn: TFS, 32, 41, 61, 119; and see also 87. Baranl's father and uncle: ibid., 25, 39. 60,127.

~ tion to Amir Khusraw and to the latter's friend and fellow-poet Amir Hasan Dihlawi, with both of whom

he claims to have been on close terms.38 Occasionally he also cites other informants, including otherwise

unknown notables who had served Balaban.39 But the assertion of this seventy-four-year-old author that

from the reign of Jalal al-Din Khalji onwards he is reliant on what he himself had witnessed (he would have

been six at Jalal al-Din's accession and twelve when the sultan died) hardly inspires confidence.40 He

possibly drew some of his information from the boon-companions (nudama') of Jalal al-Din, including

Amir Arslan *Kalahi, whom he describes as expert in history and in the practices of kings (adab-i muluk);41

but this is by no means certain. In the circumstances, it is reassuring that Barani and 'Isami, who wrote independently of each other, frequently agree in their outline of events, so that they may at least have

drawn on a common folk memory. Shades of similar traditions also appear in the brief history of Delhi

which Ibn Battuta incorporated in his travelogue.

If Juzjani has a tendency to bemuse the reader through a wealth of sometimes contradictory detail, the

problems attached to Barani's work are of a different order. Although the Ta'rikh-i Firuz-Shahi was

intended as a sequel to Juzjanl's Tabaqat42 it is in some respects inferior to it, containing as it does

relatively few dates (and some of those inaccurate) and at times describing events in a vague and

impressionistic fashion. The author himself calls his work an epitome (ijaz-u ikhtisar) and denies aiming at

completeness.43 On the other hand, he attempts what none of our other sources remotely approaches,

namely an explanation of events and policies, which in itself has raised acute problems of interpretation.44

With regard to Balaban's reign, for example, the reader is struck by a laudable attention to analysis and

characterization. The former ghulam of Iltutmish who now supplanted his master's dynasty is portrayed by

Barani as a grim ruler who was determined to be more than merely primus inter pares.45 He consciously

sought to distance the sovereign behind a screen of increased pomp and ceremony, employed a network of

spies and informers to monitor the activities of his amirs, and destroyed a number of his former colleagues

among the aristocracy. This stickler for etiquette would not even allow his private attendants to see him

without his jacket (yakta).46Himself a parvenu, Balaban is said to have refused to promote men of low

origins

38 Ibid., 67, 68,113, 183,360. 39 Khwaja Taj al-Din Makrani: ibid., 36. Khwaja Dhaki, a nephew of Balaban's wazir Basri: ibid., 114.

Qadi Sharaf al-Din *Barmas (?): ibid., 168 (printed text has SRPA'YN, but cf. BL ms., fol. 90b).

40 Ibid., 175. For Barani's age when writing, see ibid., 573.

41 Ibid., 199. 42 Ibid., 20-1. 43 Ibid., 361.

44 See, for instance, P. Hardy, Historians of medieval India (London, 1960), chap. 2; Harbans Mukhia,

Historians and historiography during the reign of Akbar (New Delhi, 1976), 3-5, 10-11, 19-26; I. Habib,

'Baranl's theory'.

45 TFS, 30, 34-5. 46 Ibid., 33, 34-5, 40.

~ and constantly to have stressed the need to restrict the ranks of the aristocracy to those of noble birth. The

impression of an intimate portrait that is all too seldom found in medieval chronicles is reinforced by a

number of speeches reportedly made by the sultan in conversation with his maliks or his sons. We might

feel ourselves to be holding the keys to a veritable treasure-house of Balaban's own policies and political

theory, and this is reflected in modern historiography. Indeed, Balaban has been hailed as 'perhaps the only

sultan of Delhi who is reported to have discussed at length his views about kingship'.47 There are grounds,

however, for approaching such reported speech with considerable reserve. Dr Peter Hardy has

demonstrated that the views expressed in these sections are those of Barani himself and are to be found

also, but more conspicuously, in his Fatawa-yi Jahandari (written some time in the 1350s), a handbook of advice for sultans set squarely in the Persian Furstenspiegel tradition.48

The Ghiyathid era

The circumstances surrounding Nasir al-Din Mahmud Shah's fate in 664/ 126649 are especially problematic.

The earliest report that Balaban murdered Mahmud Shah occurs, in fact, in Wassaf's history of India; Ibn

Battuta heard a similar story three or four decades later, and it is found also in 'Isami's Futuh al-Salatln.50

On the other hand, Barani makes no reference to foul play, and Sirhindi expressly claims that Mahmud

Shah died a natural death.51 It may well be, therefore, that Balaban has been unjustly maligned; although it

must be said that none of these sources - whether or not it charges Balaban with regicide - tells us what

became of the sons Mahmud Shah is known to have fathered. The prince whose birth to Balaban's daughter

in 657/1259 is greeted in such effusive terms by Juzjani would surely have been regarded as the future sultan.52 It is possible that the na'ib

47 Nizami, Some aspects, 280. For similar views, see Sir Wolseley Haig, in The Cambridge history of India,

III, Turks and Afghans (Cambridge, 1928), 74-5; Habibullah, Foundation, 162-3, 179; Aziz Ahmad,

Political history, 259-63, 267-71.

48 P. Hardy, 'The oratio recta of Baranl's Ta'nkh-i Firuz Shahi - fact or fiction?', BSOAS 20 (1957), 315-21.

49 The date 11 Jumada I 664/18 February 1266 is given in TMS, 39, and supported by Mahmud Shah's

coins, which go down to 664. But it should be noted that Sirhindi gives the duration of the reign as 19

years, 3 months and 16 days: this would place the sultan's death in 663/1265. In any event, the years 662 and 665 supplied respectively by TFS, 25, and by FS, 163, 164 (tr. 290, 291), are wrong.

50 Wassaf, 310; JT, ed. Jahn, Indiengeschichte, Pers. text Taf. 22, Ar. text Taf. 57 (German tr. 48). FS, 163

(tr. 289-90). IB, III, 170, 174 (tr. Gibb, 632, 635).

51 TMS, 39. For two opposing views on Mahmud Shah's death, see Habibullah, Foundation, 161, who

argues that murder is improbable, given Balaban's position and his previous relations with the sultan; and

K. A. Nizami, 'Balaban the regicide', in his Studies in medieval Indian history (Aligarh, 1956), 48-62, and

in HN, 274-5.

52 TN, I, 496 (tr. 714). As he is not named, we do not know which of the four sons listed at I. 475 (tr. 672),

was Balaban's grandson. Raverty's insertion of'the late' after each name is not justified on the evidence of

the best mss.

~ was satisfied with the prospect of his grandson's succession but that the boy died in infancy, precipitating

a crisis which was resolved by the elimination of Mahmud Shah and his issue by other unions. We can only

speculate.

In contrast with Iltutmish, Balaban was blest with two able adult sons: Muhammad, who held Sind

until his untimely death in battle with the Mongols in 683/1285 (and hence was known as Khan-i Shahid,

'the Martyr Prince'); and Mahmud, entitled Bughra Khan, who was appointed governor of Lakhnawti. But

like his old master he lost a promising heir and was followed by a frivolous youth. On the old sultan's death in 685/1287, a party headed by the influential castellan (kotwal) of Delhi, Fakhr al-Din, who had been on

bad terms with the 'Martyr Prince', ignored the claims both of Muhammad's son Kaykhusraw and of Bughra

Khan in the east, and enthroned the latter's hedonistic son Kayqubad. Their opponents, such as the wazir

Basri, suffered dismissal and exile.53 Kaykhusraw was subsequently murdered; Bughra Khan, who assumed

the style of Sultan Nasir al-DIn and advanced westwards to challenge Kayqubad, was reconciled with his

son at a meeting on the banks of the river Sarju (the episode commemorated in Khusraw's Qiran al-

Sa'dayn). Accepting the fait accompli, he restricted his ambitions to Bengal, which remained an indepen-

dent sultanate until 724/1324.

The young sultan, who moved his residence to Kilokhri, a few miles away, celebrated his freedom

from the restraint of his grandfather's reign by giving himself up to pleasure and leaving the affairs of state to the powerful dadbeg (arnir-i dad) Nizam al-Din. An able but unscrupulous man, Nizam al-Din profited

from Kayqubad's unconcern about the affairs of state to bring down the wazir and the great nobles of the

previous reign, and then induced the sultan to sanction the murder of his cousin Kaykhusraw, who had

made the elementary mistake of seeking Mongol assistance. Eventually Kayqubad tired of the dadbeg and

had him poisoned.54 Nizam al-Din's role is a difficult one to assess; it is noteworthy that Sirhindi's account

mentions him only in passing and makes no allusion to his paramountcy. But for Barani the execution or

exile of the chief men of Balaban's reign, followed by the sultan's illness and deposition in favour of his son

Kayumarth, undermined the regime: there was rivalry among the maliks, with none strong enough to

triumph. The Khalaj amir Jalal al-Din rallied his followers, seized control of Kayumarth and became na'ib;

after a short interval he set aside the infant sultan and occupied the throne himself in Rabf II 689/April-May

1290. The helpless

53 TFS, 122 (and cf. 107). TMS, 52. FS, 184-6, 196 (tr. 315-16, 328). An echo of Fakhr al-DTn's role is

found in the slightly garbled tale picked up by IB, III, 175-6 (tr. Gibb, 635-6), where he is referred to

correctly as malik al-umara but also, in error, as na'ib.

54 TFS, 170. The account in FS, 198-200 (tr. 330-2), where Nizam al-DIn is made to drink poison he had

prepared for Kayqubad, reads like the stuff of romance. For Kaykhusraw, see ibid., 196-8 (tr. 328-30).

~ Kayqubad had not long survived his deposition, dying on 19 Muharram 689/1 February 1290. In

Sirhindi's version, he simply perishes of starvation and neglect; according to another tradition, however,

Kayqubad was murdered on Jalal al-Din's orders by an officer whose father he had executed. In the Ta'nkh-

i Firuz-Shahi events move inexorably towards the overthrow of the Ghiyathids and the transfer of power to the Khaljis. Thus Bughra Khan, after the reconciliation, is said to have told his attendants that he would

never see his son again and to have prophesied the imminent downfall of Balaban's dynasty.55

Given these forebodings, the portrayal of Jalal al-Din Firuz Shah Khalji (689-95/1290-6) comes as

something of a surprise - indeed, an anti-climax. In Baram's view, kings had to balance the opposing

qualities of benevolence and severity that are necessary if kingship is truly to be a lieutenancy (khilafat,

niyabai) on behalf of God.56 It is clear that, for him, Jalal al-Din did not embody this balance. This

seasoned warrior, who prior to his accession had spent many years fighting the Mongols on the western

frontiers of the Sultanate, is written off as a pious, mild and merciful ruler who shrank from conflict that

would cost the lives of Muslim soldiers and was reluctant to shed the blood of his opponents; even thags ('thugs') captured in Delhi were shipped off down the Ganges towards Lakhnawti.57 The sultan pardoned

alike Balaban's nephew, Malik Chhajju, who rose in revolt against him in 689/1290, and a group of nobles

who had engaged in a half-hearted plot against him slightly later.58 In the speeches put into the conspirators'

mouths by Barani, they are made to criticize Jalal al-Din as unworthy of the sovereignty; it is not unlikely

that his clemency towards Chhajju's adherents outraged those who had severed their ties with the old

dynasty. But Jalal al-Din reacted differently towards the dervish Sidi Muwallih, whose hospice (khanaqah)

had become the centre of another aristocratic conspiracy and whose death at the sultan's instigation is seen

by both Barani and Tsami as presaging the collapse of the regime.59 If Jalal al-DIn's downfall, however,

was divine retribution for his treatment of a Muslim holy man, it came about more immediately because of

his childlike trust in, and indulgence towards, his scheming nephew 'Ala' al-Din, who murdered him at a

meeting on the banks of the Ganges on 16 Ramadan 695/18 July 1296 and seized the throne. Yet the old

sultan was not altogether negligible. On Barani's own testimony, Jalal al-Din headed expeditions against the Hindu kingdoms of Rajasthan, and halted an

55 TFS, 150, 156.

56 For a summary of Barani's views, see Peter Hardy, 'Didactic historical writing in Indian Islam: Ziya al-

Din Barani's treatment of the reign of Sultan Muhammad Tughluq (1324-1351)', in Yohanan Friedmann

(ed.), Islam in Asia, I, South Asia (Jerusalem, 1984). 41-4.

57 TFS, 186, 213; 189 for the thags, on whom see Hodivala, Studies, I, 266-7.

58 TFS, 190-2. TMS, 64-5. 59 TFS, 208, 212. FS, 217 (tr. 382).

~ invasion by a Mongol prince who withdrew without a battle. Khusraw's Miftah al-Futuh, written only

twelve months into the reign, reveals that even within that time Jalal al-Din also campaigned in the sub-

Himalaya against both the Mongols and the Hindus, in addition to suppressing a major insurrection by

adherents of the Ghiyathid dynasty. He conveys the impression of remarkable energy on the part of the

sultan.

It is worth comparing Barani's view of Jalal al-Din with his perspective on Balaban. During the

first few years of his reign, Balaban led an expedition to Lahore and the Salt Range (Kuh-i Jud) and

engaged in campaigns against both the turbulent Meos (Mlwat) in the vicinity of the capital and the

unsubdued infidels of Katehr, east of Bada'un. Thereafter, apart from his long march to Bengal to crush its

rebellious governor, Toghril, he does not seem to have taken the field in person. It is noteworthy that the task of repelling the Mongols was left to his sons and other lieutenants. There are hints that such apparent

sluggishness underlay the widespread desertions to Toghril not only in Bengal but even from Delhi

following the early defeat of Balaban's generals.60 BaranI evidently sees it as his duty to explain Balaban's

failure to prosecute the war against the infidel, and he does so by staging an exchange between the sultan

and some of his fellow Shamsis. Urged to undertake plundering campaigns far afield in Hindu territory in

the manner of Aybeg and Iltutmish, Balaban is made to justify his policy: caution was vital because the

Mongols were now launching annual raids on India and it was no longer possible, as it had been in bygone

days, to leave the capital and embark on distant enterprises.61 At first sight, this might appear to furnish a

persuasive rationale for the sultan's relatively unadventurous policy after c. 1270; but whether we can in

fact take it as a reflection of Balaban's own views is open to serious doubt (see below, p. 253). Thus the

contrast between Balaban, the 'strong' ruler whose energies were somewhat muted, and Jalal al-DIn, the weakling who was nevertheless strikingly active, presents us with something of a paradox.

Kingship, stability and hereditary succession

The period of sixty-two lunar years that separates the death of Iltutmish on 20 Sha'ban 633/29 April 1236

from the accession of 'Ala' al-Din Muhammad Shah Khalji in 695/1296 witnessed the reigns of ten sultans.

Those of Iltutmish's immediate successors - Firuz Shah (633-4/1236), Radiyya (634-7/1236-40), Bahram

Shah (637-9/1240-2) and Mas'ud Shah (639-44/ 1242-6) - were particularly ephemeral; Kayqubad (685-

9/1287-90), the latter's son Kayumarth (689/1290), Jalal al-DIn Khalji (689-95/1290-6) and his son Rukn

al-Din Ibrahim (695/1296) each alike enjoyed authority 60TFS, 83, 84. 61Ibid., 50-1.

~ for only a brief period. The longest reign is that of Balaban himself (664-85/1266-87), closely followed

by that of the last Shamsid, Nasir al-Din Mahmud Shah (644-64/1246-66): when these are subtracted from

the total, the average reign occupies less than three lunar years. Of the ten sovereigns, only Balaban is

known with certainty to have died a natural death. His predecessor's fate is obscure (above, p. 52), but the

others died violently, in all cases but one at the instigation, or at least following the accession, of the ruler

who replaced them; the exception, Radiyya, at the hands of Hindus in the wake of a failed bid to oust her

successor.

There does not appear to have been an accepted rule of succession, and the role played by designation was extremely limited: in fact, with the possible exception of the founder of the Sultanate

(above, p. 46), no sultan prior to Ghiyath al-Din Tughluq Shah (d. 724/1324) was succeeded by his

designated heir. As far as we can tell, none of the Shamsids was given the opportunity to nominate a

successor: it is not even known whether Radiyya, Bahram Shah or Mas'tid Shah left any issue or whether

Mahmud Shah was survived by any of his sons. Nor were Balaban's preparations for the succession

attended by better fortune than those of his Shamsid predecessors, since Kayqubad, as we have seen, was

not his heir. Kayqubad and his child were within a few years supplanted by Jalal al-Din, whose own sons

were disinherited in 695/1296 by his nephew and murderer, 'Ala' al-Din.

For much of the thirteenth century, therefore, the history of the Sultanate hardly seems to be characterized

by the essentials of stable government and might not suggest that the hereditary principle carried much weight. But if the succession failed to observe any logical pattern, it cannot be said, even so, that heredity

was immaterial. On the contrary: connections both with the present and with past ruling dynasties seem to

have been of some moment. The attempt by Iltutmish's ghulam 'Izz al-Din Balaban (later Kiishlu Khan) to

have himself proclaimed sultan following Bahram Shah's overthrow in 639/1242 was thwarted by the

prompt action of a group of his colleagues, who gathered solemnly at their master's tomb and ensured that

the throne stayed within Iltutmish's family: Izz al-Din had to acquiesce, and the choice fell on 'Ala' al-DIn

Mas'ud Shah.62 It might well be asked how, if loyalty to Iltutmish's dynasty was so strong, Ulugh Khan

Balaban was able to justify his displacement of Iltutmish's heirs. To this we can return no sure answer.

What Balaban did in 664/1266 was essentially what 'Izz al-Din had attempted to do, but he had

undoubtedly spent a longer time entrenching himself at the centre. There are grounds for believing that

Balaban was married to a daughter of Iltutmish (below). He had, moreover, a claim which was denied to

'Izz al-Din. It was thought - or Juzjani, writing in 658/1260, Wanted it to be thought - that Balaban sprang from the ruling

62 TN, II, 36 (tr. 780).

~ line of khans of Iltutmish's own clan, the Olberli (p. 63 below).63 This conceivably formed part of the

propaganda deployed in Balaban's interest when the time came to supplant the Shamsids only a few years

later.

Yet the legitimizing properties of Shamsid blood did not fade even under subsequent dynasties.

Amir Khusraw makes Mu'izz al-Din Kayqubad boast to his father of his descent not only from Sultan

Balaban but from Iltutmish and from Nasir al-Din Mahmud Shah (whose daughter was his mother).64 The sultans were naturally unwilling to tolerate the forging of such links by others. One reason why Jalal al-Din

Khalji reacted so harshly to the conspiracy to enthrone the dervish Sidi Muwallih may have been that the

latter's supporters planned to marry him to a daughter of 'Sultan Nasir al-Din' (whether the Shamsid

Mahmud Shah or the Ghiyathid Bughra Khan is not made clear).65

And when, a few years later, during the

absence of 'Ala' al-Din Khalji on campaign at Ranthanbor (700/1301), a party in Delhi seized their

opportunity to revolt and instal a dervish as sultan, Barani considers it worthy of notice that this cipher was

Iltutmish's maternal grandson.66 One of 'Ala' al-Din Khalji's concerns when he forbade his maliks to form

relationships (qarabatha) without his consent was surely to prevent them cementing unions with older

royal lines.67

What, at a juridical level, constituted a sultan's title to rule? Sources for the thirteenth century give

some prominence to the inauguration of a new reign by a pledge of allegiance (bay'at).68 This is first

mentioned in 634/ 1236, when according to JuzjanI the Turkish amirs who abandoned Firuz Shah entered

the capital and performed the bay'at to Radiyya. On her deposition in 637/1240 the maliks and amirs made

a 'general act of allegiance' (bay'at-i 'dmm) to Bahrain Shah and to the newly created viceroy (na'ib)

Aytegin in the royal quarter (dawlatkhana) in Delhi. Juzjani, whose phrasing suggests that he may himself

have participated in the ceremony, says that it was attended by 'the maliks, amirs, 'ulama', sadrs and the

leading figures both in the military and the capital (akabir-i lashgar-u hadrat)\ On the news that Lahore

had fallen to the Mongols in 639/1241, Bahrain Shah took the precaution of having the bay'-at repeated by

'the

63 Ibid., II, 43, 45, 47-8 (tr. 791, 796, 799-800), and Juzjani's verses in praise of Balaban, calling him 'khan of the Ilbari and king (shah) of the Yemek', at II, 220-1 (tr. 1295); cf. the information about Iltutmish's

father at I, 441 (tr. 599). P. B. Golden, 'Cumanica II. The Olberli (Olperli): the fortunes and misfortunes of

an Inner Asian nomadic clan', AEMA 6 (1986 [1988]), 27-8. On the Yemek or Kimek, a Turkish people

who had presided over a loose confederacy of tribes in the Irtysh region until its disintegration in the

eleventh century under pressure from the Qipchaq, see Hudud al-Alam, tr. Minorsky, 99-100 and notes at

304-10; P. Pelliot and L. Hambis, Histoire des campagnes de Gengis Khan. Cheng-wou Ts'in-tcheng-lou

(Leiden, 1951, vol. I only), 95-6; Golden, 'The peoples of the south Russian steppe', in D. Sinor (ed.), The

Cambridge history of early Inner Asia (Cambridge, 1990), 277-80. Nizami (HN, 251-2 n.40) is (no doubt

rightly) dismissive of the story of Balaban's royal ancestry.

64 QS, 22, 118. HN, 307. 65 TFS, 210-11. 66 Ibid, 279. 67 Ibid, 286. 68 E. Tyan, 'Bay'a', Enc.Isl2

~ people of the city' (khalq-i shahr). The oath must therefore have been taken by the leading Muslim

citizens of Delhi, who, as we shall see, were still being termed khalq ('the people' par excellence) by

Barani, over a century ater. This widening of the circle of persons from whom the pledge was equired set a

precedent for the following reign, for at the accession of 'Ala' il-Din Mas'ud Shah, the amirs, we are told,

'administered to the people a public act of homage' (khalqra baycat-i 'amm dadand).69

Juzjanl provides the fullest description of the bay'at in connection with he accession of Nasir al-Din

Mahmud Shah in 644/1246, and shows that here were in fact two ceremonies, involving respectively the

grandees and he citizens of Delhi:

The maliks, amirs, sadrs, grandees (kubra'), sayyids (sadat) and 'ulama' hastened to he exalted court and

attained the kissing of the blessed hand of that emperor shahanshah) ... Each, as befitted his status (hal),

offered congratulations on his accession. And on Tuesday the 25th [of Muharram] he held a general

audience in the ball of the Kushk-i Firuzi in the fort (qasr) of the Dawlatkhana; and they administered to all

the people (khalq) a general oath of allegiance (bay'at-i 'amm) to recognize] the sovereignty and to obey the

edicts of that... monarch.70

Although we have less information about the bay'at given to Mahmud Shah's successors, it appears to have

followed a similar pattern, for Ibn Battuta was told that the oath to Kayqubad was taken first of all by the

Malik al-Umara' (the kotwal Fakhr al-Din), then by the amirs and principal officers, and the next morning by 'the rest of the people' (i.e. of Delhi).71 From Balaban's reign, at least, a new sultan was expected to order

the release of prisoners, a practice still observed at the time of Qutb al-DIn Mubarak Shah's accession in

716/1316.72

The extension of the bay'at reflects the growing importance of the leading Muslim citizens of old Delhi,

who had played some part in the accession of Radiyya and who must have given Bahram Shah considerable

support to enable him to withstand a siege in the capital for almost three months in 639/1242.73 Nor did

their capacity to influence events cease when Kayqubad transferred his residence to Kilokhri, a few miles

closer to the Yamuna and referred to as 'the new town' (shahr-i naw) in 658/1260.74

In 689/1290 their

attachment to the Ghiyathid dynasty and hostility to the new regime would prevent Jalal al-DIn Khalji from installing himself in Delhi

69 TN, I, 456, 463, 466, 468, and II, 23 (tr. 636, 649, 656, 661, 750-1).

70 Ibid., I, 477 (tr. 675-6 modified).

71 IB, III, 176-7 (tr. Gibb, 636). TMS, 39-40, 53. For the bavcat to Jalal al-DIn at Kilokhri, seeTFS, 181.

72 Amir Hasan Dihlawi, cIshq-Nama (700/1301), tr. M. I. Borah, 'A short account of an unpublished

romantic masnavi of Amir Hasan Dihlavi', NIA 2 (1939-40), 260. TFS, 339, 382. FS, 354 (tr. 551-2).

73 7W, I, 456, 467 (tr. 635-6, 659). IB, III, 166-7 (tr. Gibb, 631). HN, 241-2. 74 For Kilokhri, see TN, II, 83 (tr. 856-7); the earliest mention is at I, 456 (tr. 634, 636).

~for some time.75 At times, too, they would fall foul of their sultan. Following an abortive revolt in

700/1301, 'Ala" al-Din Khalji conceived an aversion for the notables of Delhi. Many sadrs were banished,

and the sultan would not enter the city but took up residence instead in the suburbs ('imranat); it may have

been partly for this reason that he afterwards fortified Siri and made it his headquarters.76 Later there are

reports of antipathy between the people of Delhi and Muhammad b. Tughluq (p. 165 below).

The historians of the Delhi Sultanate still await as yet the techniques of literary analysis adopted by

Marilyn Robinson Waldman in her monograph on the Ghaznawid chronicler Bayhaqi.77 But it has been pointed out that they move on a different plane from those who now use their writings. They (and perhaps

Barani in particular) sought to reflect an ideal temporal order, in which the world is governed jointly by

pious scholars and pious sultans, and one in which change is intelligible in terms not of the human actions

the historians themselves narrate, but of divine providence.78 Certainly the verdicts of a Juzjani or a Barani

may reveal as much about what was expected of a ruler as about real personalities. It was necessary to

dispense justice to one's subjects and to supervise the affairs of state in person; to endow charitable Islamic

foundations and to treat with respect the 'ulama' and other members of the religious class, virtues for which

even the tyrannical Shah Terken is praised79 and in which Bahrain Shah was notably deficient. Nor was

mildness necessarily a virtue in a sultan. Rukn al-Din Firuz Shah's clemency and humanity (hilm-u

muruwwa) attract favourable comment, but his reluctance to injure another human being is expressly

presented as the cause of his downfall; and Mas'ud Shah's merciful treatment of his uncles, whom he

released from confinement, ultimately provided the amirs with a serviceable alternative to his rule.80 The monarch had to know when to act harshly and when to show mercy, thus avoiding the extremes of either

Balaban or Jalal al-Din. A sultan's addiction to pleasure is frequently depicted as conducive to chaos, and

an antipathy towards luxury, pomp and display, as evinced by Bahrain Shah, was

75 TFS, 172, 173.

76 Ibid., 283. FS, 277 (tr. 453), confirms his resentment. For 'Ala' al-DIn and Siri, see Lai, History of the

Khaljis, 326.

77 M. W. Waldman, Towards a theory of historical narrative: a case-study of Perso-Islamicate

historiography (Columbus, Ohio, 1980) (cf. also E. A. Poliakova, 'The development of a literary canon in medieval Persian chronicles", the triumph of etiquette', Iranian Studies 17 (1984), 237-56). But see Peter

Hardy, 'Approaches to pre-modern Indo-Muslim historical writing: some reconsiderations in 1990-1', in

Peter Robb (ed.), Society and ideology. Essays in South Asian history presented to Professor K. A.

Ballhatchet (Oxford and Delhi, 1993), 49-71.

78 Peter Hardy, 'The Muslim historians of the Delhi Sultanate: is what they say really what they mean?',

JASP 9 (1964), part 1, 59-63; also his 'Force and violence', esp. 196-204.

79

TN, I, 454 (tr. 630-1). 80

Ibid., I, 454, 457, 470 (tr. 630, 637, 664-5).

~ praiseworthy.81 So was generosity, provided that it was directed towards those who mattered (and not to

lowborn favourites, as was the munificence of Firuz Shah, Mas'ud Shah and Kayqubad). We might also

observe, perhaps, that it was vital to cherish the maliks and amirs, including those inherited from one's

predecessor: the Shamsid era and the reign of Kayqubad both furnished cautionary tales about the fate of

sultans who disregarded this last precept.

81 Ibid., I, 462 (tr. 649).

~CHAPTER 4

Turks, Tajiks and Khalaj1

Turks and military slavery

Tabaqa 22 of Juzjani's work comprises biographies of twenty-five Shams! ghulams. Although the

chronicler does not specify slave status in every case, his usage of the word 'Turk' suggests that for him it

had come to denote simply a Turkish slave (see appendix I). Already, during Iltutmish's reign, a few of

these amirs had been granted Turkish titles that included the element khan - not borne, it should be noted,

by Ghuri or Tajik notables and thus representing an innovation.2 But a significant proportion of the twenty-

five attained high office only some time after their master's death. The future Sultan Balaban, as Juzjani's

own patron and viceroy (na'ib) to the reigning monarch, receives the longest biography. The list of ghulams

represented by the biographies is also, of course, far from exhaustive; both here and elsewhere in the

Tabaqat other slaves of Ututmish, who are not accorded biographies of their own, are brought to our notice.

The pronounced slant of tabaqa 22 towards Turkish slave officers serves to obscure an important

fact. At no point did Turkish ghulams enjoy the monopoly of rank and office that they seem to have

exercised in Mamluk Egypt. One important difference was the opportunities for advancement available to

the offspring of ghulams in the Delhi Sultanate. This was not the case in Egypt, where the sons of mamluks

- the awlad al-nas - were deliberately excluded from the highest positions in the state.3 In India Turkish

ghulams also had to share power with other, non-servile groups. These included not only free Turkish

nobles, Khalaj, Ghuris, Tajiks and (from Balaban's reign) Mongols, but also other slave elements, both

black

1 This chapter is a greatly expanded version of my 'The Mamluk institution in early Muslim India', JRAS

(1990), 340-58. 2 I. Habib, 'Formation', 11 and n.62.

3 D. Ayalon, 'Studies on the structure of the Mamluk army - II', BSOAS 15 (1953), 456-8, repr. in his

Studies on the Mamluks of Egypt (1250-1517) (London, 1977); see also his 'Awlad al-nas', Enc.Isl2.

~African (Habashi, literally 'Abyssinian')4 and Indian. Although JuzjanI mentions Hindu infantrymen,

palks,5 as serving in Muslim campaigns, it is not until Balaban's reign that we read of them forming a royal

guard; and they came to play a more prominent role only in the Khalji era. Afghan troops, lastly, were part

of the military establishment of the thirteenth-century Sultanate, though appearing only fitfully in the

sources.6

It is impossible to document the training of the Sultanate's Turkish slaves, as has been done for

Mamluk Egypt, or to compose a survey of the slave contingents, of the kind that Professor Edmund

Bosworth has produced for the Ghaznawids.7 As we might guess even without Juzjani's occasional

references, the accomplishments especially valued were equestrian skills and marksmanship.8 But other

skills were not unknown, for Aybeg had received instruction from his first master in reciting the Qur'an and

was accordingly known as Qur'an-khwan.9 The sources do not usually tell us at what point a slave was

manumitted. JuzjanI alleges that on Mu'izz al-Din's death both Aybeg and Yildiz requested manumission

from the new sultan of Ghur. According to the same author Iltutmish had even prior to this been freed by

Aybeg on Mucizz al-Din's express instructions, and Ibn Battuta later heard a story that he showed his deed of manumission to the jurists of Delhi when he became sultan.10 We learn from Barani alone that Balaban

had been freed at some point prior to his accession.11 Slaves of the reigning sultan bore the designation

'Sultani'.12 Whether or not there was a recognizable cursus honorum is unclear.

The information we are given concerning the twenty-five Shamsi slaves reveals diverse ethnic and

geographical origins. Only one was apparently an Indian - Hindu Khan, who may have ranked as the major-

domo in overall charge of the sultan's ghulams, since JuzjanI says that he bore the style of mihtar-i

mubdrak and that he stood in the relation of a father to his fellow-Shamsls.13 The Turkish ghulams included

Rumis (presumably Greeks or Slavs from Byzantine territory)14 and 'Khita'Is' (Khitan from northern

4 C. F. Beckingham, 'Habash, Habasha, iii', Enc.Is2.

5 Sir Henry Yule and A. C. Burnell, Hobson-Jobson: a glossary of colloquial Anglo-Indian words and

phrases, new edn W. Crooke (London, 1903), 748-9, 'pyke, paik .

6 TN, II, 80 (tr. 852). TS, IOL Persian ms. 412, fol. 52 (extracts tr. in Mirza, Life and works, 51-2). gS, 47.

TFS, 58.

7 Hassanein Rabie, 'The training of the Mamluk Faris', in Parry and Yapp, War, technology and society,

153-63. C. E. Bosworth, 'Ghaznevid military organization', Der Islam 36 (1960), 40-50; also idem,

Ghaznavids, 101-6.

8 TN, I, 416 (tr. 513), for Aybeg; I, 443 (tr. 604-5), on the exploits of Iltutmish in battle with the Khokhars; II, 27 (tr. 756), for *Kirit Khan.

9 SA, 21. 10 TN, I, 373, 444 (tr. 398, 605). IB, III, 164 (tr. Gibb, 629-30). 1' TFS, 25, azad shuda (and pace

Nizami, in HN, 281 ).

12 HN, 224. Aziz Ahmad, 'The early Turkish nucleus in India', Turcica 9 (1977), 101, 102. wrongly

assumes that the suffix denotes immigrants of free status.

13 TN, II, 18-19 (tr. 744-6). For the position of mihtar-i saral at the Ghaznawid court, see Bosworth,

Ghaznavids, 104.

14 ‘Izz al-Din Kablr Khan Ayaz (ayaz, 'clear', 'cloudless': Sauvaget, 'Noms et surnoms', no

~ China), whose ethnic background may or may not distinguish them from the Qarakhita'is (i.e. Qara-

Khitan).15 Several of the Shamsis belonged to the Qipchaq, the group of tribes which occupied the steppes

north of the Black Sea and the Caspian.16 And particular mention should be made, lastly, of those who

belonged to Iltutmish's own people, the Olberli, a subgroup of the Qipchaq (or possibly of the Qangli, who

were closely related to them): they included Baha' al-Din Balaban, the future sultan, known as Balaban-i

Khwurd ('the Lesser').17

Although the Shamsis included a few former ghulams of other rulers,18 most were obtained direct

from slave traders: Ibn Battuta heard much later that Iltutmish as sultan sent merchants to Samarqand,

Bukhara and Tirmid to buy Turkish slaves on his behalf.19 The date of purchase ranged over a considerable period, beginning when Iltutmish was muqtac of Baran.20 The avenues varied by which Turkish youths

destined for Egypt and Syria came into the hands of slave traders,21 and the same must be true of Muslim

India. Iltutmish himself had allegedly been sold into slavery by his envious brothers, which enabled Juzjani

to liken him to the Patriarch Yusuf (Qur'an, sura 12:7-20).22

Of Sayf al-Din Aybeg (later dadbeg), it is said

that he was- enslaved 'through the perversity of kindred'. Two others were

36), Badr al-Din Sonqur (sonqur, 'gerfalcon': ibid, no. 22), and another Badr al-DIn Sonqur who would

later obtain the title Nusrat Khan: TN, II, 5, 24, 42 (tr. 724, 752, 787).

15 KhitaiIs: Sayf al-Din Aybeg, nicknamed Yaghantut ('seize elephant[s]'), and Sayf al-DIn Ikit Khan Aybeg-i Khita'I, ibid., II, 9, 28 (tr. 731, 757). Qarakhita'is: 'Izz al-Din Toghril toghan Khan (toghan,

'falcon': Sauvaget, no. 140), Ikhtiyar al-Din Aytegin Qaraqush Khan (aytegin, 'moon-prince': ibid., no. 41;

qaraqush, 'eagle': Clauson, Etymological dictionary, 670) and another Ikhtiyar al-DIn Aytegin (later the

first ghulam to hold the office of na'ib): TN, II, 13, 19, 22 (tr. 736, 746, 749). For the title Ikit Khan, see

below, p. 73, n.76.

16 Qamar al-Din Qiran Temiir Khan (qiran, 'one who slaughters': Sauvaget, 'Noms et surnoms', no. 182);

Taj al-Din Sanjar (sanjar, 'one who pierces': ibid, no. 107), nicknamed qabaqulaq ('of the protruding ears':

see Clauson, Etymological dictionary, 580-1, 621, and Jackson, 'Mamluk institution', 342 n.7); Taj al-Din

Sanjar *Kirit Khan; Ikhtiyar al-Din Yuzbeg Toghril Khan; 'Izz al-Din Balaban (later to be styled Kiishlti

Khan); and Sayf al-Din Aybeg Shamsl-yi 'AjamI: 7W, II, 17, 25, 27, 30, 36, 40 (tr. 742, 754, 756, 761, 775,

788-9). 17 To distinguish him from 'Izz al-Din Balaban (see preceding note, and below). The others were his brother

Sayf al-Din Aybeg (later entitled Kishli Khan); and their cousin Nusrat al-Din Sanjar (Shir Khan): Raverty

read Shir Khan's personal name as Sonqur, but BL ms., fol. 211a, and IOL ms., fol. 291b, read SNJR. For

the ascription of Iltutmish and these ghulams to the Olberli ('LBRY in Hablbl's edition), see TN, I, 440, 441,

and II, 43, 45, 47 (tr. 598, 599, 791, 796, 800); also Golden, 'Cumanica II. The Olberli'. On the Qipchaq-

Qangli relationship, see Pelliot and Hambis, Histoire des campagnes, 95-116; Hudud al- Alam, tr.

Minorsky, 304-10; C. E. Bosworth, 'Kanghli', Enc.hr.

18 cIzz al-Din Kablr Khan Ayaz, bought from the family of Yildiz's amir-i shikar, Naslr al-Din Aytemur al-

Baha'i, so called because he had belonged to Baha' al-Din Toghril; and Nusrat al-Din *TaIsI, the one-time slave of Mucizz al-Din himself: TN, II, 5, 7, 10 (tr. 724-5, 727, 732). The meaning of Taisi's name, given

consistently as TAYSY in BL ms. (fols. 182b, 199b, 200b, 202a, 218a), is unknown.

19 IB, III, 171 (tr. Gibb, 633). 20 TN, II, 4 (tr. 723).

21 D. Ayalon, 'Mamluk', Enc.Isl2, VI, 314.

22 TN, I, 441 (tr. 599-600); and cf. the remarks about Yusuf (Joseph) at I, 439 (tr. 596-7).

~ rumoured to be of Muslim parentage and thus unlawfully enslaved.23 Kishli Khan is said to have been

enslaved when young, having fallen into Mongol hands.24 From the 1220s the westward advance of the

Mongols gave rise to a sharp increase in the supply of Turkish slaves, particularly from the Caspian and Pontic steppes. Unscrupulous rulers seized on those who sought asylum with them, like the Turkish

chieftain in the Crimea who in 640/1242-3 sold the future Mamluk Sultan Baybars into slavery;25 desperate

fugitives exchanged their own offspring for the necessities of life; and the conquerors themselves converted

human booty into more liquid assets by unloading their able-bodied captives onto the market. Iltutmish

may also have profited from internal convulsions among the stricken Olberli.26

The attractions of an elite corps of military slaves who possessed no local ties and whose sole

loyalty was to the master who had bought, nurtured and trained (and sometimes manumitted) them are

obvious. A number of authors, including the litterateurs Jahiz in the ninth century and Ibn Hassul in the

eleventh, and the Seljukid wazir Nizam al-Mulk (d. 485/1092), had sung the praises of Turkish ghulams.27

At the beginning of the thirteenth century Fakhr-i Mudabbir (admittedly writing for a monarch who was himself a ghulam) was the latest in a long line of authors to do so. There is no kind of infidel people, he

says,

which is brought over to Islam and does not look with longing at home, mother, father, and kindred: for a

time they are bound to adopt Islam, but in most cases they apostatize and relapse into paganism. The

exception is the Turkish race, who, when they are brought over to Islam, fix their hearts in Islam so firmly

that they no longer remember home or region or kinsfolk ... The Turk is like a pearl that lies in the oyster in

the sea. For as long as it is in its habitat, it is devoid of power and worth;

23 Sayf al-Din: phrase omitted in Habibi's edition, ibid, II, 41, but cf. BL ms., fol. 211a, ba- inad-i aqriba (also Raverty's tr., 790). Muslim parentage: TN, II, 24, 33-4 (tr. 752, 766).

24 Ibid., II, 45 (tr. 796).

25 Peter Thorau, The lion of Egypt: Sultan Baybars I and the Near East in the thirteenth century, tr. P. M.

Holt (London, 1992), 28. Al-Yunini, al-Dhayl 'ala Mir'ati'l-Zaman (Hyderabad, AP, 1374-81/1954-61, 4

vols.), Ill, 240.

26 TN, II, 45 (tr. 796), Golden, 'Cumanica II. The Olberli', 28; and Thomas T. Allsen, 'Prelude to the

Western campaigns: Mongol military operations in the Volga-Ural region, 1217-1237', AEMA 3 (1983), 5-

24 (esp. 16).

27 C. T. Harley Walker, 'Jahiz of Basra to al-Fath ibn Khaqan on the "Exploits of the Turks and the army of

the Khalifate in general"', JRAS (1915), 631-97 (esp. 662 ff., 682, 685). Ibn Hassul, Risala, ed. Abbas

Azzawi and tr. serefeddin Yaltkaya, 'Ibni Hassul'un Turkler hakkinda bir eseri', Belleten 4 (1940), Ar. text

40-3, Tu. tr. 259-61 (I owe this reference to the kindness of Dr Carole Hillenbrand). Extracts from this last

passage are translated in D. Ayalon, 'The Mamluks of the Seljuks: Islam's military might at the crossroads',

JRAS, 3rd series, 6 (1996), 314-15. See further ibid., 316-19; and for further references, idem. 'Aspects of

the Mamluk phenomenon, I. The importance of the Mamluk institution', Der Islam 53 (1976), 212-16, repr.

in his Mamluk military society; Andre Wink, 'India and Central Asia: the coming of the Turks in the

eleventh century', in A. W. Van den Hoek et al. (eds.), Ritual, state and history in South Asia. Essays in

honour of J. C. Heesterman (Leiden. 1992), 764-5.

~ but when it emerges from the oyster and from the sea, it acquires value and becomes precious, decorating

the crown of kings and adorning the neck and ears of brides.28

This is not to say, however, that contemporaries were oblivious of the Turk's limitations. In one of

'Awfi's anecdotes Iltutmish deliberately chooses a Tajik to investigate an officer's financial interests, a

delicate task for which, we are told, the 'impetuosity' (tahawwur) of a Turk would have disqualified him.29

And it is a moot question how deeply Islam was ingrained in these first-generation converts. If Turkish

slaves may have enjoyed the benefits of being reared as orthodox Muslims, their origins lay, nevertheless,

in the pagan steppelands of Central and Western Asia. This is not the place to examine the question of

pagan survivals within Muslim Turkish societies.30 But Radiyya's enthronement may be symptomatic.

Although the accession of a female monarch (as opposed to a regent) was without precedent in the Islamic

world, the list of Qara-Khitan sovereigns in the twelfth century furnishes two examples. Some of Iltutmish's ghulams belonged, as we saw, to the Khitan or the Qara-Khitan, and in general women in the eastern steppe

enjoyed greater freedom.31 It may well be that in raising up their master's daughter Turkish officers were

strongly influenced by their pagan background.

The problem of the Chihilganis

Although Barani's Ta'nkh-i Firuz-Shahi opens with Sultan Balaban's accession, he prefaces his account of

the reign with some remarks about Balaban's predecessors. They are very brief, but they do at least

endeavour to make sense of the Shamsid era. In Iltutmish's time, he says,

illustrious maliks and amirs ... and many wazirs and notables (ma'arif) came to the court of Sultan Shams al-Din [Iltutmish] from fear of the slaughter and terror of the accursed Mongol Chingiz Khan ... But after

the death of Sultan Shams al-Din his Turkish chihilgani slaves grew powerful. The sons of Sultan Shams

al-Din ... were unable to fulfil the duties of kingship ... and as a consequence of the ascendancy of the

Turkish Shamsi slaves all those great men of high birth ... were destroyed on every pretext during the reigns

of Sultan Shams al-Din's sons, who had no notion of the world or about rulership. And following the

elimination of those grandees and commanders, the Shamsi slaves rose and became khans. Every one of

them attained

28

SA, 35-7. 29

JH, BL ms. Or. 2676, fol. 263b.

30 J. P. Roux, 'Recherche des survivances pre-islamiques dans les textes turcs musulmans: le Babar-Name',

JA 256 (1968), 247-61, and 'Recherche des survivances pre-islamiques dans les textes turcs musulmans: le

Kitab-i Dede Qorqut, JA 264 (1976), 35-55.

31 The point is made by Habibullah, 'Sultanah Raziah [sic]', IHQ 16 (1940), 752, though his other examples

comprise female regents. See further my 'Sultan Radiyya bint Iltutmish', in Gavin R. G. Hambly (ed.),

Women in the medieval Islamic world: power, patronage, piety (New York, 1998), 181-97. On the Qara-

Khitan, see Karl A. Wittfogel and Feng Chia-sheng, History of Chinese society: Liao 907-1125

(Philadelphia, 1949), 643, 644, 646; also the remarks about the Liao rulers of China ibid., 199-202. Juzjani

was aware of only one female Qara-Khitan monarch: 7W, II, 95-6 (tr. 911).

~ new riches, palaces, pomp and magnificence ... Because the Shamsi slaves were of one master (khwajatash buda), and all forty became great at one time, one did not bow before another or obey him, and

they demanded equality in iqta's, troops, high rank and honour ... As a result of the inexperience of

Iltutmish's sons and the supremacy of the Shamsi slaves, the monarchy had forfeited all majesty.32

Who were the Chihilganls? This question was investigated in a stimulating article by Professor Gavin

Hambly, who reached no definite conclusion as to the origin or meaning of a term not used by Juzjani or, in

fact, in any Indian source other than Barani's work.33 It is true that at one point above Barani employs

instead the term 'forty' (chihil), which led the sixteenth-century compilators Harawl and Firishta to assume

that Iltutmish had forty slaves: this in turn induced modern historians to speak of a 'college' of forty.34 Yet

the concept is of dubious value. On every other occasion Barani has recourse to the distributive numeral,

which strongly suggests that the Chihiganis were so termed because each commanded a corps of forty ghulams. It is worth noting that in contemporary Egypt there were amirs commanding units of forty royal

mamluks; we should perhaps conclude that the Chihilganls formed a parallel group of commanders within

the ranks of Iltutmish's Shamsi slaves.35 As Hambly observes, Barani ascribes only three amirs by name to

the ranks of the Chihilganls;36 it is worth noting that they are all relatively junior ghulams of Iltutmish.

The rise of the Shamsi ghulams

The bloody conflict outlined by BaranI is nowhere mentioned explicitly by JuzjanI, writing when the

hegemony of the Shamsl ghulams was at its zenith; but its onset is clearly visible in his account of the

turbulent era of Iltutmish's heirs. In all likelihood Firuz Shah, who according to 'Isami failed to accord his

father's Turkish slaves sufficient attention,37 relied excessively upon a number of Tajik bureaucrats whom

the Turkish slaves 32 TFS, 27-8; for a fuller translation of the passage, see I. Habib, 'Formation', 15-16. There is a brief

reference to this phase of the Sultanate's history at TFS, 550.

33 Gavin R. G. Hambly, 'Who were the ChihilganI, the forty slaves of Sultan Shams al-Din Iltutmish of

Delhi?', Iran 10 (1972), 57-62. For an alternative view, see Khurram Qadir, 'The amiran-i-chihalgan of

northern India', JCA 4, no. 2 (1981), 59-146.

34 Nizam al-Din Ahmad Harawi, Tabaqat-i Akbari, ed. B. De (Calcutta, 1931-5, 3 vols.), I, 78 (tr. 93), and

Firishta, I, 130. Both were possibly influenced by 'Isami's story that Iltutmish was offered forty slaves and

bought them all except Balaban, the future sultan: FS, 122 (tr. 238). For the 'college', see Haig in Cambridge history of India, III, 61-2; Habibullah, Foundation, 346; Nizami, Some aspects, 111 n.7, and in

HN, 232-4. I. Habib, 'Formation'. 16, takes 'forty' in a less literal sense.

35

Ayalon, 'Studies on the structure of the Mamluk army - II', 469-70.

36 Ulugh Khan Balaban, his cousin Shir Khan, and Temiir Khan (later muqta' of Samana and Sunnam after

Shir Khan's death: below, p. 77): TFS, 25, 65; Hambly, 'Who were the ChihilganI?', 61.

37

FS, 130 (tr. 248, but n.l ibid, is misleading, since it cites as examples nobles who were not slaves).

~ massacred at Tara'in in the course of the sultan's campaign against the rebel Kabir Khan and his allies.

Radiyya, by contrast, was vigorously supported by her father's Turkish ghulams (umara-yi turk ki

bandagan-i Shamsi budand).38 But she soon began to construct a power-base of her own. When the Turk

Sayf al-Din Aybeg-i *Tutuq, whom she had appointed as her deputy in command of the army (na'ib-i

lashgar) with the style of Qutlugh Khan, died in 635/1237, his office passed not to a Turk but to the Ghuri

amir Qutb al-DIn Hasan b. 'A1I.39 She was deposed because in promoting an African (Habashi) slave,

Jamal al-Din Yaqut, to the rank of intendant of the stable (amir-i akhiir) she had alienated the 'Turkish

maliks and amirs who were Iltutmish's slaves' and in particular the amir-hajib, Ikhtiyar al-Din Aytegin.40 A

rising by Kabir Khan at Lahore in 636/1239 failed, but in the next year Aytegin and his ally *Altunapa, the

governor of Tabarhindh, contrived a mutiny while the sultan was on campaign, and Yaqut was executed;

Radiyya was incarcerated at Tabarhindh under the supervision of * Altunapa.41

With the enthronement of Mu'izz al-Din Bahram Shah (637-9/1240-2), the Turkish amirs took

further steps to concentrate power in their own hands, with the formal institution of the office of na'ib

(viceroy), which was conferred on the amir-hajib Aytegin; it is significant that their oath of allegiance

(bay'at) to the new sovereign was conditional upon Aytegin's appointment.42 But when Aytegin usurped

certain imperial prerogatives, Bahram Shah grew resentful of his tutelage and had him murdered in

Muharram 638/July 1240; the office of na'ib lapsed.43 For a short time the direction of affairs was in the

hands of another Shamsl, the new amir-hajib Badr al-Din Sonqur-i Rumi. Sonqur rendered the sultan

valuable service in the campaign against * Altunapa, who had reacted to the elimination of his ally Aytegin

by marrying Radiyya and marching on Delhi to restore her to the throne.44 The principal role, however, was

passing to Junaydi's successor as wazir, Nizam al-Mulk Muhadhdhab al-Din, who fell out with Sonqur and

poisoned the sultan's mind against him. When Sonqur hatched 38 TN, II, 36 (tr. 779); at I, 458 (tr. 640), they are called simply umara-yi turk.

39 Ibid., I, 459 (tr. 641-2). Sayf al-Din's sobriquet, given as BHTW in Habibi's text and as 'Bihaq' by Raverty,

appears as TTQ in BL ms., fols. 182b, 183a. This looks like the Tu. title tutuqltotaq, on which see Denis

Sinor, 'The Turkish title tutuq rehabilitated', in Turcica et Orientalia. Studies in honor of Gunnar Jarring

(Istanbul, 1988), 145-8; alternatively it could be a nickname, tutuq, 'tongue-tied' (Clauson, Etymological

dictionary, 453), or 'lip' (Sau-vaget, 'Noms et surnoms', no. 124). For his epitaph, from Abuhar, see ARIE

(1970-1), 18-19, 119 (no. 4).

40 HN, 240, 243. TN, II, 21, 22-3 (tr. 748, 750); BL ms., fol. 183a, gives Yaqut the title 'chief amir' (amir

al-umara"), a phrase omitted in the printed text of TN (I, 460). 41 See generally Habibullah, Foundation, 119-21. *Altunapa's name is spelled 'LTWNYH in the printed

text (Raverty's 'AltQnlah'), but I suspect that ya is an error for pa or ba and that we have here Tu. altun,

'golden', + abalapa, 'ancestor', or oba, 'clan', 'tribe', found among the Qipchaq/Polovtsy: Pol'noe sobranie

russkikh letopisei, I. Lavrent'evskaia letopis', 2nd edn (Leningrad, 1926-8), col. 278; Clauson, 5-6, 131.

42 TN, I , 463 (tr. 649). 43 Habibullah, Foundation, 121-2. M TN, II, 24 (tr. 753).

~ a conspiracy to replace Bahram Shah with one of his brothers, the wazir reported it and Sonqur was

banished from court to his iqta of Bada'un in Safar 639/August 1241. Returning without permission three

months later, he was imprisoned and put to death.45

Bahram Shah in turn was overthrown in 639/1242 when, under the influence of one of his

courtiers, Fakhr al-Din Mubarak Shah Farrukhi, he contemplated the wholesale removal of the Turkish

slave officers.46

In Jumada I 640/October 1242 the Turkish commanders attacked and killed the wazir

Muhadhdhab al-Din, who had played them off against Bahram Shah and who now sought tio concentrate

power in his own hands and to exclude the Turkish amirs from all state business; he seems to have been the

last wazir with military inclinations for almost a century. The fact that the ringleaders were not punished

but were in fact rewarded suggests that the new sultan, Mas'ud Shah, was behind them.47 Like his two

predecessors, however, Mas'tid Shah tried to cut the Turkish amirs down to size. Although Juzjani does not

define the 'nobodies' (nakasan) who had wielded influence at court during the final months of his regime, a later account suggests that he relied on black African slaves.48

Thus far, then, Juzjani and other authors do provide corroborating evidence for Barani's analysis. There are

grounds, nevertheless, for regarding that analysis as deficient in two respects. At no time, firstly, did a party

comprising Turkish ghulams exclude free elements, whether Turks or not. Opposition to Rukn al-Din Firuz

Shah (and then initially to Radiyya) brought together the Turkish ghulam Kabir Khan, the free Turkish

noble 'Ala' al-Din Jam, the Ghurl amir Salari, and the presumably Tajik Junaydi.49 Several Tajik officials

were implicated in Badr al-Din Sonqur's plot to remove Bahram Shah: among them were the chief qadi,

Jalal al-Din Kasani, who was deposed and banished from Delhi, and the accountant-general (mushrif-i

mamalik) Taj al-Din Musawi, who was executed with Sonqur in 639/1241.50 Prior to his execution, Sonqur

vainly sought the protection of the Ghurl amir Qutb al-Din Hasan.51 Juzjani assures us that Ghurl and Tajik

as well as Turkish maliks were affronted at the position of Yaqut in Radiyya's counsels; and of Bahram Shah we learn, again, that he

45 Habibullah, Foundation, 122-3.

46 TN, I, 466-7, and II, 20, 30 (tr. 658-9, 747, 761-2).

47 Ibid., I, 469, and II, 27, 42 (tr. 662, 757, 787). Taj al-Din Sanjar *Kirit Khan was promoted to the rank of

intendant of the imperial elephants (shihna-yi pit) and subsequently to that of sar-i jandar, while Badr al-

Din Sonqur Sufi-yi Rumi (the future Nusrat Khan, not to be confused with his namesake above) took over

the dead wazir's territory of Kol.

48 Ibid., I, 471 (tr. 668-9). FS, 144. TMS, 34.

49 TN, I, 455-6, 458 (tr. 633-4, 639). It is noteworthy that Kabir Khan and Jam had suffered a lapse from

favour during Iltutmish's latter years and that Salari, who had served the late sultan as amlr-hajib, is not so

described under the new reign. Jan! and Kabir Khan: ibid, II, 6, 9 (tr. 726, 731-2). Salari: Habib,

'Formation', 13; also JH, I, 12.

50 TN, I, 464-5 (tr. 652-3, 654). 51 Ibid., II, 25 (tr. 753).

~aroused the fears of Ghtiri as well as Turkish amirs.52 The rejection of 'Izz al-Din Balaban and the choice

of Mas'ud Shah similarly demonstrate an alliance of different elements. The notion of the sovereignty

passing to one of Iltutmish's Turkish ghulams perhaps found little favour with the Ghuris, while the other Shamsis for their part were unwilling to jettison the family of their old master. It has been rightly pointed

out that the structure of power that emerged in 639/1242 bears the marks of a compromise among the

various groups within the elite.53 The office of na'ib was revived and conferred on Qutb al-Din Hasan; a

senior Shamsi, Qaraqush Khan, was made amir-hajib; and Taj al-Din Sanjar-i Qabaqulaq, one of the three

amirs said to have checked Tzz al-Din Balaban's pretensions, received the iqtac of Bada'un.54 Collaboration

between Turk and non-Turk was evidently not beyond the bounds of possibility. It appears, however, that

what made the internal crisis in the Sultanate so protracted, and so dangerous, was a split among the

Shamsis themselves.

The second defect of Barani's analysis is that it treats the Shamsis as a monolithic group.

Historians of the parallel Mamluk military slave institution in Egypt and the Near East are accustomed to speaking of khushda-shiyya, the sense of comradeship and unity of interest that bound together the slaves

of the same master. Such sentiments, however, often failed to outlive the master himself, and to pay too

much attention to khushdashiyya is to court the risk of over-simplification.55 It is surely possible - though

the sources do not reveal it - that individuals among Iltutmish's elite corps of ghulams were conscious of

closer links with colleagues from the same tribal background. What is still more likely is that there was

initially a stronger sense of solidarity among the junior ghulams, who would have been a distinct group

with interests of their own.

We should note how many of the Turkish slaves who were instrumental in Firuz Shah's downfall

and Radiyya's accession seem to have been junior ghulams still employed in the imperial household. A group described by Juzjani as 'the Turks of the court' (or 'the capital') had manifested their disenchantment

with Firuz Shah at an early stage by leaving Delhi for 'Hindustan', conceivably in order to join his brother

in Awadh. But they were intercepted; among them was Balaban 'the Lesser', who suffered a brief spell of

imprisonment.56 It was 'the Turkish amirs and personal slaves who were serving in the centre' (umara-yi

turk-u bandagan-i khass ki dar khidmat-i qalb budand) who had mutinied at Tara'in; and these same

officers, called now 'the centre [consisting] of Turkish amirs' (qalb-i umard-

52 Ibid., II, 22-3, 164 (tr. 750, 1133). 53 Habibullah, Foundation, 24.

54 TN, I, 468 (tr. 661-2); cf. also II, 20, 26, 36-7 (tr. 747, 755, 780).

55 D. Ayalon, 'L'esclavage du mamelouk', IONS 1 (1951), 29-31, 34-7, repr. in his Mamluk military society; cf. also the remarks of Robert Irwin, The Middle East in the middle ages: the early Mamluk Sultanate

1250-1382 (London and Sydney, 1986), 65, 88-90.

56 TN, II, 48-9, 51 (tr. 802, 805).

~ yi turk), who deserted Firuz Shah at Kilokhri and recognized Radiyya.57 In all probability Iltutmish had

purchased many of them at a relatively recent date. 'Izz al-Din Balaban (Kiishlu Khan), for example, one of

the two men named as the ringleaders at Tara'in, had been acquired during the siege of Mandor (i.e. in

624/1227); by the time of Iltutmish's death he had become muqta of Baran.58 But the emeute surely

involved many others who now received important offices at court or their first iqta's as a reward for

bringing Radiyya to the throne. *Altunapa, to whom she transferred Baran, had been Iltutmish's chief canopy-bearer (sar-i chatrdar).59 Baha' al-Din Balaban 'the Lesser', purchased by Iltutmish only in

630/1232-3, was at his master's death merely a falconer (khasadar); Radiyya promoted him to amir-i

shikar.60 Of his brother Sayf al-Din Aybeg (the future Kishli Khan), purchased in the course of an embassy

from Iltutmish to Egypt and Baghdad which can be reliably dated to 629/1231-2, we are told that until

Radiyya's accession he had simply served in the sultan's private household (khidmat-i dargah-i khass

mikard); but he now became deputy commander of the guard (nd'ib-i sar-i jdndar). Taj al-Din Sanjar (later

Arslan Khan), obtained from the same source and probably around the same time, was like Balaban a

falconer; but Radiyya made him cupbearer (chashnigir) and subsequently allotted him the iqta' of

Balaram.61 The status and aspirations of such ghulams would have set them not only against outsiders -

including free Turkish grandees - but even, on occasions, against Iltutmish's more senior slaves like Kabir

Khan, who had long ago attained high rank in the state apparatus and received an iqta'. The history of

Iltutmish's successors is in large measure the story of the rise of his junior slaves to positions of power and of the tensions among them that threatened to tear the infant Delhi state asunder.

Balaban and his rivals

Balaban 'the Lesser', who under Bahrain Shah had been promoted from amir-i shikar to amir-i akhur, had

distinguished himself in the siege of Delhi in 639/1242, for which he received the iqta' of Hansi. Since he is

known to have enjoyed the patronage of the late amlr-hdjib Badr al-Din Sonqur, who

57 Ibid., I, 456 (tr. 634-5, 636).

58 Ibid., II, 36 (tr. 777-9 garbled). Kushlu[k] means 'strong', 'powerful': TMENP, III, 639 (no. 1676). 59 TN, II, 21 (tr. 748).

60

Ibid., II, 48, 51 (tr. 802, 806); for the date of purchase, II, 48 (tr. 801). The meaning of khdsadar was

established by Hodivala, Studies, II, 67-8.

61 Kishli Khan: TN, II, 45, 46 (tr. 796-8). Arslan Khan: ibid, II, 33, 34 (tr. 766 and n. 3, 767); for arslan,

'lion', see Sauvaget, 'Noms et surnoms', no. 4. Embassy from Delhi to Egypt in 629/1231-2: Ibn al-

Dawadari, VII, 305. Kishli, which may mean 'humanity' (Clauson, Etymological dictionary, 754), is found

in the title of a Khwarazmian amir earlier in the century: TJG, I, 80 (tr. 103, mistakenly equating it with kiishlu).

~had secured for him his first iqta' at Rewari, he doubtless participated in the revolt against Bahrain Shah

from a desire to avenge his old friend. The same circumstance may have led him to share also in the attack

on Sonqur's enemy Muhadhdhab al-Din, since Juzjani's phrasing suggests a link between the wazir's death

and Balaban's promotion to amur-hajib.62 This was at the expense of Qaraqush Khan, who was dismissed

to his iqta' of Bhayana. It is difficult to know what to make of his subsequent transfer from Bhayana to

Kara or the bald statement at the end of his biography that in 644/1246, following the overthrow of Mas'ud

Shah, he was killed in that region.63 All this might indicate that he was a rival of Balaban, who had possibly

engineered his demotion; but we cannot be sure. At any rate, Balaban was almost certainly instrumental in

Mas'ud Shah's removal and the enthronement of Mahmud Shah (644-64/1246-66), events related by Juzjani

in highly anodyne terms.

In 647/1249 Balaban became na'ib and was granted the style of Ulugh Khan, and the sultan married his

daughter. His allies among the nobility were also favoured. The new viceroy transmitted his office of amir-

hdjib to his brother Sayf al-DIn Aybeg, now styled Kishli Khan, and a number of other supporters were

promoted: the Shamsi Taj al-Din *Teniz Khan, who is invariably described as a faithful henchman of

Balaban, became deputy amir-hajib; Balaban's own slave, Ikhtiyar al-DIn Aytegin-i mui-yi daraz ('the

long-haired'), hitherto deputy to the amir-i akhur, moved up to succeed Kishli Khan in that office.64 The

wazir Sadr al-Mulk Najm al-Din Abu Bakr, who had succeeded Muhadhdhab al-DIn around the time of

Balaban's own appointment as amir-hdjib, appears to have been another ally.65

From about this juncture we begin to discern the dim outline of an opposition group, also led by Shamsis. Already, we are told, Balaban's promotion to the dignity of amir-hajib had been resented by other maliks.66

The new na'ib and his allies proceeded to make a concerted attack on 'Izz al-Din Balaban, who in 639/1242

had been consoled with the style of Kushlii Khan and an extensive but distant iqta' of Nagawr and had since

643/1246 held the additional grant of Multan. Dissatisfied with this, he obtained from Mahmud Shah

Uchch as well, on condition that he relinquish Nagawr; but he failed to honour his part of the bargain.

Multan, which Kiishlii Khan lost to Hasan Qarluq's forces, was subsequently occupied by Ulugh Khan

Balaban's cousin Shir Khan, from whom Kushlu

62 TN, I, 469, and II, 51-2 (tr. 663-4, 806-7); but cf. II, 53 (tr. 809), where Balaban's appointment alone is

mentioned and is dated in 642/1244-5.

63 Ibid., II, 20 (tr. 747). 64 Ibid., II, 59-60 (tr. 820-1); cf. also II, 29, 46 (tr. 759, 798). *Teniz Khan's title figures in Habibi's text as

TR; Raverty reads 'Tiz', and in BL ms., fol. 206, while there is no dot above the middle 'tooth', the final

letter is clearly z. For Tu. tenizldeniz, 'sea', 'ocean', see Clauson, Etymological dictionary, 527; TMENP, III,

205-7 (no. 1192). That Aytegin-i Mui-yi Daraz was Balaban's slave we know from TFS, 83.

65 TN, I, 469 (tr. 663-4). Habibullah, Foundation, 126. 66 TN, II, 52 (tr. 807).

~ Khan vainly endeavoured to take it. After a campaign headed by Ulugh Khan and the sultan ousted him

from Nagawr, which was conferred on the na'ib's brother Kishli Khan, Kushlu Khan retired to Uchch,

where he was taken prisoner by Shir Khan and released only after ordering the garrison to surrender. Kushlu Khan, who had thus been deprived of all his iqta's in favour of the viceroy's supporters and

kinsmen, was compensated with Bada'un early in 649/1251.67

Kushlu Khan had his revenge during a campaign to the north-west in 650-1/1252-3, when Ulugh Khan

Balaban was dismissed first to Hansi and then, deprived of Hansi in favour of an infant son of the sultan, to

Nagawr. He was replaced as na'ib by Qutb al-Din Hasan, and in a general reshuffle of appointments his

friends and family were demoted. Kishli Khan, *Teniz Khan and the wazir Sadr al-Mulk were all removed

from office; Juzjani for the second time forfeited his post of chief qadi; and Shir Khan, whose extensive

iqta's comprised Uchch, Multan and Tabarhindh, was dislodged by the sultan's forces and retired into

Mongol territory. Kushlu Khan and his allies, who included the shadowy Qutlugh Khan and the Indian 'Imad al-Din Rayhan, shared out offices and iqta's among themselves: Rayhan became wakil-i dar, and

Kushlu Khan recovered his old iqta's in Sind.68

Juzjani tells us frustratingly little about Ulugh Khan Balaban's enemies. Although he devotes a fair-sized

biography to Kushlu Khan, no such compliment is paid to Qutlugh Khan, who was sufficiently important to

marry the sultan's mother.69 To label the opposition to Balaban as an 'anti-Turkish' faction70 is to be misled

by Juzjani s polemic contrasting 'Turks' and 'Hindus'. By his own admission they included Kushlu Khan

and Qutlugh Khan as well as lesser figures like the latter's son-in-law 'Izz al-Din Balaban-i Yiizbegi; and he

specifically mentions Turks who were allied with Rayhan out of hostility to Ulugh Khan.71 Juzjani writes

bitingly of Rayhan

67 Ibid., II, 37-8 (tr. 780-4); see also I, 484-5, and II, 44, 46, 61-2 (tr. 689-90, 792, 798, 822-3). 68 Shir Khan: ibid., I, 487 (but reading az masdff-i kunar-i db-i Sindh for the az masaff-i kuffar-i ab-i Sindh

of the text), and II, 34 (tr. 695-6, 767); see also II, 38 (tr. 784) and 44 (to be corrected from IOL ms., fol.

291a; Raverty's tr., 792, garbled). FS, 146-9 (tr. 269-74), seems to have a distorted account of this

campaign, allegedly against the Mongols and dated 656. The year, which Habibullah (Foundation, 136)

puts even later, is impossible, since Qutb al-DIn Hasan (d. 653/1255) is listed among the commanders and

Balaban-i Zar (i.e. Kushlu Khan) is left at Uchch and Multan on the sultan's return. For Shir Khan's flight to

the Mongols, see p. Ill below; for the rest, Habibullah, Foundation, 126; he had wrongly assumed (ibid.,

125) that Qutb al-DIn had not survived 'Ala' al-Din Mascud Shah.

69 TN, I, 489 (tr.,701); cf. also I, 493, and II, 39 (tr. 710, 785). S. B. P. Nigam, Nobility under the Sultans of Delhi A.D. 1206-1398 (Delhi, 1968), 40, believes that the marriage, which he dates at the onset of 1255,

alienated the sultan; but we have no evidence as to when it took place.

70 Habibullah, Foundation, 126, 132, 195. Nizami, Some aspects, 141, and in HN, 262. Cf. also P. Saran,

'Politics and personalities in the reign of Nasir-ud-Din Mahmud the slave [sic]', in his Studies in medieval

Indian history (Delhi, 1952), 228.

71 TN, II, 68 (tr. 833); cf. II, 70 (tr. 836), for 'Turks' and 'Hindus', but also a reference to

~ as a baseborn Indian eunuch (majbub-u naqis-u az qaba'il-i Hind); this suggests, incidentally, that he was

of slave status and renders it unlikely that he belonged to what could properly be termed an emerging

Indian Muslim noble class. Rayhan's candidate for the office of chief qadi, Shams al-Din, hailed from Bahraich, and the iqta' of Bahraich is later said to have been restored (ruju1 shuda) to Rayhan on his

dismissal from court in 653/1255.72 Since Bahraich had been Mahmud Shah's iqta' prior to his accession,

there is a strong possibility that Ulugh Khan's enemies drew support from the sultan's own power-base and

that Rayhan was the sultan's own slave; Mahmud Shah himself, as well as his mother and her husband, was

doubtless behind them.

Balaban and his followers regained power by dint of allying with the sultan's renegade brother Jalal al-DIn

Mas'ud, who six years previously had fled from his iqta's by way of Santur to take refuge with the

Mongols.73 JuzjanI is reticent concerning his subsequent activities, and we are dependent instead on the

history of India presented by Wassaf and Rashid al-Din, writing in Mongol Persia.74 According to their

version of events, Jalal al-Din had grown apprehensive of the hostility of a number of Iltutmish's old slaves. Although the Iranian tradition does not offer a wholly reliable guide to the history of the Sultanate and

various details are incorrect,75 these slaves can, with one exception (Qutlugh Khan), be identified with

persons named by JuzjanI as allies of Ulugh Khan Balaban.76 Jalal al-Din Mas'ud's return with a Mongol

army, and the creation of a client state for him around Lahore and the north-western Panjab, will be dealt

with later

Rayhan's association with Qutlugh Khan. For Yiizbegi, see ibid., I, 487 and II, 64 (tr. 695, 827).

72

Ibid., II, 66 (tr. 829), for a brief notice of Rayhan; I, 487, and II, 64 (tr. 694, 827), for Shams al-DIn; II,

69, for Rayhan and Bahraich. Nigam, Nobility, 39 n.37, also links Rayhan with Bahraich. 73 TN, I, 482 (tr. 683-4). For Santur, see Hodivala, Studies, I, 229.

74 Wassaf, 310; JT, ed. Jahn, Indiengeschichte, Ar. text Taf. 57, Pers. text Taf. 21-22 (German tr. 47-8);

Qashanl (c. 1318), Ta'rikh-i Uljaitu Sultan, ed. Mahin Hambly (Tehran, 1348 Sh./ 1969), 184-5.

75 The late date given for the prince's flight, 651/1253-4, is probably that of his reappearance in India with a

Mongol army. Another error is that Jalal al-DIn is said to have been an earlier sultan, deposed by Ulugh

Khan in favour of Radiyya. Qutlugh Khan is listed among Ulugh Khan's confederates. If this is not an error,

then the two men may have become enemies only after Jalal al-Din's departure; Wassaf's account does in

fact claim that Qutlugh Khan subsequently grew fearful of Ulugh Khan.

76 See Hodivala, Studies, II, 78. 'Sungur Khan' is very probably Shir Khan Sanjar. Aybeg-i Khita'I, muqtac

of Baran, bore the style of *Ikit Khan: Raverty gives his style as 'Ban' or 'Bat' Khan, but cf. BL ms., fols.

192b, 194a, and IOL ms., fols. 262a, 263b (also Habibfs apparatus, I, 488, 491). Sayf al-DIn Aybeg-i

Shamsi cAjamI, whom JuzjanI entitles 'Erkli Dadbeg' (Tu. erkli, 'having authority', 'one's own master':

Clauson, Etymological dictionary, 224), had held the office of aniir-i dad since 640/1242-3. 'Yuzbak' is

Ikhtiyar al-DIn Yiizbeg Toghril Khan (yiizbeg, 'commander of a hundred': ibid., 983), who had apparently

replaced Jalal al-DIn himself as muqtac of Qinnawj and who was also surely a member of the na'ib's

affinity, since just prior to the Qinnawj grant Balaban had been instrumental in his restoration to the sultan's

favour.

~ (p. 111) When in 652/1254 he advanced east from Lahore, he was joined by a number of amirs, headed by Ulugh Khan, who had lost out in the power struggle of the previous year.77 Inconclusive manoeuvres by

the rebels and the sultan's army were followed by a compromise of which Rayhan was the immediate

victim: he was relieved of his office and dismissed to his new iqta' of Bada'un.78 It seems that he was

discarded by the sultan and by certain of those who had earlier profited from Ulugh Khan's removal. One of

them was surely Arslan Khan, who had supplanted Shir Khan at Tabarhindh but who now appears among

Jalal al-Din Mas'ud's followers. Another was Qutlugh Khan's son-in-law, 'Izz al-Din Balaban-i Yuzbegi,

who negotiated with Jalal al-Din Mas'ud and Ulugh Khan on the sultan's behalf and who narrowly escaped

assassination by Rayhan's agents: he is found enjoying the court's favour henceforward.79 Jalal al-Din

Mas'ud, on the other hand, had reaped less from the settlement than he might have anticipated. Along with

his confederates, he was reconciled with Mahmud Shah; but we read only that Lahore was recognized as

his iqta', and he seems to have withdrawn there and ceased to play any role in events at the centre. At any

rate, there is no mention of him as accompanying the imperial army when Ulugh Khan Balaban and the sultan re-entered Delhi in Dhu'l-Hijja 652/ January 1255.80

Ulugh Khan Balaban was swift to reimpose his dominance at court. In Rabi' II 653/June 1255 the na'ib

Qutb al-Din Hasan, who appears to have attempted to mediate in the preceding struggle, was arrested and

executed, allegedly for some remark which had offended the sultan. Ulugh Khan was restored to the

viceroyalty, while the dead man's iqta' of Mirat was transferred to Kishli Khan, once more amlr-hajib.%x

Ulugh Khan had also wasted little time in moving against the opposition group. At the very beginning of

653/1255, Qutlugh Khan and the sultan's mother were dismissed from court and ordered to take up

residence in Qutlugh Khan's new iqta' of Awadh. Around the same time Rayhan was deprived of Bada'un in

favour of Ulugh Khan's adherent *Teniz Khan and transferred to the more distant Bahraich, where in Rajab

653/August 1255 he was killed by Taj al-Din Siwistani.82 Qutlugh Khan maintained the struggle for some time in Awadh, before joining forces with Kushlii Khan from Sind in 655/ 1257; attempting to manufacture

a coup in Delhi which was frustrated by

77

Including Kishli Khan from Kara and Aybeg-i Khita'i from Sunnam and Mansurpur. The unnamed amir

from Awadh, TN, II, 66 (tr. 830), is probably Yuzbeg Toghril Khan, who received that territory after being

removed from Qinnawj for insubordination by Qutb al-DIn Hasan b. 'A1I: ibid., II, 31 (tr. 762), where no

dates are given, but this seems the most plausible reconstruction of events.

78

Ibid., I, 488-9, and II, 66-8 (tr. 699-700, 830-4).

79 Ibid., II, 68, 78 (tr. 832, 833, 849). Balaban-i Yuzbegl may have continued to bask in the court's favour

as late as 656/1258 (below, p. 92).

80 TN, I, 489, and II, 68-9 (tr. 700, 834). 81 Ibid., I, 489, and II, 46 (tr. 702, 798-9). 82 Ibid., I, 489, 490, and

II, 69, 70 (tr. 701, 702-3, 834, 835-6).

~ Ulugh Khan's adherents, the allies briefly besieged the capital but were obliged to retreat following the

na'ib's arrival with his army.83 Nothing more is heard of Qutlugh Khan, who may have left India to seek

shelter with the Mongols.84 Kushlii Khan, for his part, retired to Sind: according to Juzjani, his forces were

heavily depleted, since most of the contingents from Uchch and Multan deserted him and many took

service with Ulugh Khan and the court.85 As far as we can tell from Juzjani's account of the next few years,

Ulugh Khan Balaban's opponents were excluded from any share of power at the centre; he held the viceroyalty unchallenged until his usurpation of the throne itself some ten years later.

In his analysis of the period preceding Balaban's accession as sultan, Dr Nigam sees the pattern as the

elimination of rival elements such as Africans or Tajiks, which left the Turks unchallenged, followed by a

phase in which rival Turkish factions struggled for power but in a more restrained fashion, involving

bloodless changes of regime and compromises.86 Whether the conflicts of the 1250s were in fact more

restrained, on one level, is highly questionable. Admittedly there was no repetition of the massacre of 634/

1236, which has the appearance of small-scale genocide; but we still see the political murders of

individuals like Qutb al-Din Hasan. The virulence of the struggle surprised not only contemporary

observers but even the protagonists. When Kushlu Khan was obliged in 648/1250 to go to relieve Uchch,

which was under attack from Shir Khan, he pinned his hopes, we are told, on the fact that they were both 'of one house and one threshold'.87 In other words, since the two amirs had been slaves of Iltutmish, he

anticipated that they would be able to reach some amicable arrangement. He was to be disappointed: Shir

Khan placed him in custody and released him only when the city had been taken. 'Never could there be a

more amazing case than this', exclaims Juzjani, describing how Balaban's forces and those of Kushlu Khan

and Qutlugh Khan confronted each other in 655/1257; 'for they were all alike of one purse and messmates

of one dish, between whom the accursed Satan had brought forth such discord.'88 There are grounds for

suggesting, in fact, that the situation in the 1250s was not less but more dangerous because it could not be

resolved merely by the mass disposal of a group of Tajik bureaucrats. Rather, it involved a contest between

two more nearly equal parties, who engaged in full-blown civil war. Both groups, moreover, were prepared

to seek the assistance of the pagan Mongols. Ulugh Khan Balaban and his confederates regained power by

83 Ibid., I, 490-1, 492-4, and II, 39, 69-75 (tr. 703-6, 707-10, 784-6, 835, 836-44). 84 JT, ed. Jahn, Indiengeschichte, Ar. text, Taf. 57, alone specifies that Qutlugh Khan made for Mongke's

court; the Persian text (Taf. 22), like Wassaf, 310, and Qashani, 184-5, says merely that he set out in Jalal

al-Din's wake (tr. 48 has misleadingly 'schlossen sich Galal ad-Din an'). He was allegedly accompanied by

'Sungur Khan' (i.e. Shir Khan): this is unlikely, though Shir Khan too is known to have fled to the Mongols.

85 TN, I, 493, and II, 39 (tr. 710, 786). 86 Nigam, Nobility, 37-8. 87 TN, II, 38 (tr. 783). 88 Ibid., II, 73

(Raverty's tr. 841 modified).

~ allying with a Mongol satellite; and Kushlu Khan, as we shall see, reacted to his defeat by turning Sind

into a Mongol encampment.

The Ghiyathid aristocracy

We have, unfortunately, minimal information on Balaban's assumption of the Sultanate when Mahmud

Shah died; as we saw (p. 52), even the manner of the late sultan's death and the fate of his offspring are

unclear. It is likely, though by no means certain, that the Shamsid regime was terminated by the use of open

force. Nasir al-Din Mahmud Shah is the only Shamsid monarch apart from Firuz Shah known to have had

his own establishment prior to his accession, so that he was accompanied to Delhi from Bahraich by 'great

numbers of horsemen and paiks' (mabalighi-yi mard-i paik-u suwar).s9

His possession of independent

resources of manpower may be one reason why he retained the throne for longer than the brief interval vouchsafed to other members of his dynasty, especially if he was behind the opposition to Ulugh Khan.

Ulugh Khan Balaban's long career prior to his enthronement, particularly as muqta' of Hansi since

639/1242, had furnished him with the means to acquire a force of personal retainers (hashamha-yi khass).90

It doubtless included the equipage of a thousand paik slaves that accompanied him on his hunting

expeditions as sultan and is described as being 'of long standing' (qadim).91 Before he became sultan, he

also possessed Turkish ghulams of his own. We meet them already in the pages of the Tabaqat-i Nasiri.

important and influential officers like Aytegin-i Mui-yi Daraz and the sipahsalar Qarachomaq, who

represented Ulugh Khan in the negotiations of 652/1254 with the sultan.92 Their tribal origins are unknown.

Presumably they were purchased after the final Mongol assault on the Qipchaq in 1239-40, when his own

fellow-tribesmen, the Olberli, were finally scattered and many fled across the Black Sea into Anatolia;93

like the Egyptian Sultan al-Salih Ayyub, Balaban could have profited from a glut on the market in Turkish youths.

It must also be borne in mind that his long tenure of the viceroyalty since 647/1249, with only a brief

interruption, enabled Ulugh Khan Balaban to manoeuvre his own supporters and friends into strategic

positions. His brother Kishli Khan was succeeded as amir-hajib on his death in 657/1259 by his son 'Ala'

al-Din Muhammad, who retained the office well into

89 Ibid., I, 479 (Raverty's tr. 677 unaccountably renders mard as 'domestics').

90 Ibid., II, 72, 74 (tr. 839, 841). 91 TFS, 55.

92 TN, II, 67 (tr. 831-2). His name is Tu. qara, 'black', + chomaq, 'mace' (Clauson, Etymological dictionary,

422-3; cf. Sauvaget, 'Noms et surnoms', no. 93). For Aytegin, see above, p. 71.

93 al-Yunini, III, 240. Irwin, Middle East, 17-18, calling them 'Barali'.

~ Balaban's reign.94 It is possible (though by no means guaranteed) that those who emerge as supporters of

Balaban in earlier crises backed his seizure of the throne. Nusrat Khan Badr al-Din Sonqur, for example,

who had moved swiftly to Delhi's assistance in 655/1257 when it was besieged by the na'ib's enemies, is

known to have remained a muqta' until at least 669/1271.95 In the preface to his Ghurrat al-Kamal,

moreover, Amir Khusraw contrives a word-play alluding to his maternal grandfather, the 'arid 'Imad al-

Mulk (d. c. 671/1272-3), as 'sultan-maker'; and we know from the same author that 'Imad al-Mulk's own

establishment had a complement of 200 Turkish and 2000 Hindu slaves and 1000 horsemen.96 Possibly 'Imad al-Mulk - a Shamsi according to Barani - was instrumental in Ulugh Khan Balaban's usurpation of

the throne: certainly the 'arid's department had been active in the defence of Delhi on Balaban's behalf in

655/1257.97

A later tradition claims that Balaban, as sultan, abolished the office of na'ib, although it is known to have

been revived before his death.98 Barani says that in his aim of destroying his former colleagues

(khwajatashan), the great Shamsi maliks, he had a number poisoned, so that his cousin Shir Khan, who

held the iqta's of Lahore, Sunnam and Deopalpur, would not come to court either during Mahmud Shah's

reign or in Balaban's, for fear of meeting the same fate; eventually, however, in c. 668/1269-70, Balaban

had him poisoned also. Those Shamsis who survived did so only by virtue of the sultan's own favour.99

Barani names two of them, Temur Khan and 'Adil Khan. Temiir Khan appears as Temiir Khan Sonqur-i 'Ajami, malik of Kuhram, in Juzjani's list of Mahmud Shah's nobles: after Shir Khan's death, Balaban

granted him the iqta's of Sunnam'and Samana. 'Adil Khan is surnamed Shamsl-yi 'AjamI, which suggests

his identity with the amir-i dad Sayf al-DIn Aybeg (above, n.76).100 Of the ultimate fate of these

94 TN, I, 495 (tr. 713). TFS, 35, 36-7, 113-14.

95 TN, II, 42 (tr. 787-8). G. Yazdani, 'Inscription of Sultan Balban from Bhayana, Bharatpur State', EIM

(1937-8), 5-6 (though identifying him in error with the Nusrat Khan of 'Ala' al-Din Khalji's reign).

96 DGK, IOL Persian ms. 51 (Ethe 1186), fol. 34b, agar nishan-i sultarii nadasht sultan-nishani dasht (loosely translatable as 'If he was not [himself] the sultan, he [nevertheless] made the sultan'). The phrase is

mangled in the printed edition by Sayyid Wazir al-Hasan 'Abidi (Lahore, 1975), 67, which also gives in

error a figure of 100,000 for the Hindu slaves and omits the Turks.

97 TN, I, 493 (tr. 709). For a sketch of cImad al-Mulk's career, see TFS, 114. He is possibly identical with

Iftikhar al-Mulk Sharaf al-DIn Muhammad Rashidi, who is said to have occupied the dlwdn-i 'ard-i

mamdlik when Balaban was na'ib: DA, RRL Persian ms. 1231, fol. 56b, and tr. Sh. Abdur Rashid, 'Dastur-

ul-Albab fi cIlm-il-Hisab', MIQ 1 (1950), 93. That 'Imad al-Mulk's name was Muhammad is clear from TS,

IOL Persian ms. 412, fol. 36b, where his son is addressed as 'Mahmud-i Muhammad'. The assumption that

he was of Indian extraction is not warranted by the sources.

98 DA, fol. 56b (tr. Rashid, 93). But see TMS, 51, 52, for the na'ib Kochu. 99 TFS, 47-8, 65; and a brief reference at 550.

100 Ibid., 36, 50, 65, 83. Temiir Khan in Mahmud Shah's reign: TN, I, 476, and BL ms., fol. 188b (tr. 673).

For an inscription of Aybeg-i Shamsi-yi 'Ajamfs son Muhammad at Farrukhnagar in Gurga'un, dated

674/1276, see G. Yazdani, 'The Inscriptions of the Turk

~ grandees, however, we are not informed. That Balaban's purge was by no means complete is evident from

Barani's comment on the number of sons of Shamsi ghulams who held office in the Ghiyathid era.101

By modern historians Balaban has been charged with sapping the roots of Turkish power in India.102 But his purpose, of course, in bringing down a number of his former colleagues was to promote his own slaves.

Toghril, who usurped control of the distant province of Lakhnawti and was overthrown in c. 680/1281-2, is

the most notorious of them. Balaban's favourite ghulam, according to Barani, was Ikhtiyar al-Din Begbars,

who became amir-hdjib/barbeg, possibly in succession to 'Ala' al-Din Muhammad b. Kishli Khan,

accompanied the sultan on his Bengal campaign and was given the job of hunting down Toghril when the

revolt collapsed.103 Others of whom we are told incurred the sultan's displeasure and forfeited their high

offices and even in some cases their lives, as did Malik *Buqubuq (see p. 101) and Aytegin-i Mui-yi Daraz

(Amin Khan), muqta' of Awadh, hanged in c. 678/1279-80 for his failure to suppress Toghril's revolt. The

sons of Balaban's slaves, referred to as his mawlazadagan ('sons of freedmen'), also played a significant

role in the affairs of state. An example is the sar-ijandar Ikhtiyar al-Din 'Ali b. Aybeg, popularly known as

Hatim Khan, who was granted Amroha as his iqtac early in Balaban's reign and under Kayqubad received

Awadh and the style of Khan Jahan; the poet Amir Khusraw was for some years in his service.104 In the reign of Balaban's successor there was still a recognizable and self-conscious group of 'Ghiyathis', former

ghulams of the old sultan or their offspring, like Ikhtiyar al-Din Alp Ghazi who opposed the newly

established Khalji regime in 689/1290.105

Although Balaban, therefore, changed many of the personnel, ghulam status and ancestry persisted as

qualifications for high office. Yet in the course of his reign the ruling class was certainly broadened. No

more than his Shamsid predecessors did Balaban preside over an elite that was exclusively of Turkish

origin and composed only of his close kin or his slaves and their progeny. Barani depicts him as virulently

hostile to the

Sultans of Delhi', EIM (1913-14), 26-7; RCEA, XII, 206-7 (no. 4711). In addition, Balaban's old ally Nusrat Khan was still in command at Bhayana in 669/1271 (above).

101 TFS 66 Taj al-Din, the son of Qutlugh Khan-i Shamsi, is named as one of the 'amirs of Hindustan' sent

against the rebel Toghril: ibid., 83. His father is probably the Qutlugh Khan who died in 635/1237 (above,

p. 67).

102 E.g. by Saran, 'Politics and personalities', 242-3; Nigam, Nobility, 42, saying that Balaban 'sought to

destroy the Turkish nobility'; Nizami, Some aspects, 143, and in HN, 285-6.

103 TFS 61, for his closeness to Balaban; and see also 24, 81, 88. His name, usually transliterated 'Bektars', shows clearly in BL ms., fols. 32a, 47a, as BYKBRS; for the same form in contemporary Egypt, see al-

Safadi, Waji , X, 187-8.

104 TFS, 36, 118-19; and for Awadh, QS, 221; RI, V, 55. Mawlazada is explained in Hodivala. Studies, I,

342. On Khusraw, see Mirza, Life and works, 66, 69, 70-3.

105 TMS, 63: on him, see n.128 below. For another example of an amir with this sobriquet. 'Qutlugh

Sultani Ghiyathf, see Yazdani, 'Inscriptions of the Turk Sultans', 31-2.

~ promotion of the lowborn and as refusing to appoint a certain Kamal-i Mahyar to the post of revenue-

intendant (khwaja) at Amroha. We might be less inclined to accept this testimony at face value and to

dismiss it as Baram's personal view mediated through the sultan, were not a slightly fuller version of the story found in a fifteenth-century source. But Balaban's objection was clearly based on the fact that Kamal-i

Mahyar was the son of a Hindu slave.106 His own career having been temporarily blighted by the Indian

ghulam Rayhan, the sultan could well have conceived an aversion for persons of the same background. It is

noteworthy that Balaban's antipathy did not extend to Indians - whether converts or not - of noble

extraction. We are told that his servitors included a certain 'Hatya Paik', presumably a Hindu aristocrat,

who received the high stipend of 100,000 jitals (i.e. approximately 2000 tangas).107 In the wake of his

campaign against the people of the Salt Range (Kuh-i Jud) in c. 665/1266-7, Balaban brought back with

him to Delhi the two sons of their raja, who adopted Islam. The appearance of both these princes, 'Ali Shah

Kuhijudi and 'Izz al-Din Khurram, together with the despised Kamal-i Mahyar, among the maliks of

Balaban's grandson Kayqubad may be taken to signal the rise of an Indian Muslim aristocracy even prior to

the Khalji era, with which it is traditionally associated.108

The nobility still included Tajiks, of whom the most prominent in Balaban's latter years was the kotwal of

Delhi, Fakhr al-Din, entitled 'Chief Amir' (Malik al-Umara). Barani, who claims that Fakhr al-Din and his

father had occupied this post between them for eighty years, thereby allows us to identify the father as

Jamal al-DIn Nishapturi, described as ulugh kotwalbeg in 655/1257 when Delhi was under attack by

Balaban's enemies: Nishapuri origins would explain the grandiloquent title Khan-i Khurasan which Fakhr

al-Din acquired under Kayqubad.109 His nephew and son-in-law, Nizam al-Din, who became Kayqubad's

dadbeg and who is described as a survivor 'from among the illustrious Shamsi and Balabani maliks', is a

mysterious figure and seems to emerge out of thin air, unless he is to be identified with Nizam al-DIn

Buzghala, Balaban's wakil-i dar.uo During Balaban's reign, too, Muslim notables continued to arrive from

the many territories occupied by the Mongols.111 One especially distinguished immigrant was the deposed

sultan of Kirman, Hajjaj, who remained for ten years 106 TFS, 36-7. Sayyid Ashraf Jahanglr Simnani, Maktubat-i Ashrafi, BL ms. Or. 267, fol. 66b.

107 TFS, 210.

108 TMS 40, 54. See TFS, 126, for all three among Kayqubad's nobles. I. H. Siddiqui, 'Social mobility in the

Delhi Sultanate', in I. Habib (ed.), Medieval India 1, 24.

109 Jamal al-Din: TN, I, 493 (tr. 709). Fakhr al-Din: TFS, 117-18, 126 (to be corrected from BL ms., fol.

67b); his career is outlined by B. S. Mathur, 'Malik-ul-Umara Fakhruddin -the kotwal of Delhi', IC 39

(1965), 205-8.

110 TFS 131; I68; cf. also 24, 36, 37, 191. Buzghala means 'kid' in Persian: see the word-play in

TS, IOL Persian ms. 412, fol. 40a. 111

Firishta, I, 131, citing 'Ayn al-Din Bljapuri.

~ and left for his homeland only after the accession of Jalal al-Din Khalji, but died en route at Bhakkar

towards the end of 690/1291.112

It appears that Balaban may have consciously built up the power of Khalaj amirs and profited from the

influx of Mongol notables following the upheavals in Mongol territory after c. 1260 (below, pp. 108-10, 115-16). His brother Kishli Khan, as amir-hajib in the 1250s, is said to have been on good terms with the

Khalaj amirs, among others, while at a later date Khalaj officers served in Sind under Balaban's son

Muhammad, the 'Martyr Prince'.113 Amir Jamal Khaljl, whom Balaban created na'ib-i dadbeg, was

probably his hajib Jamal al-Din 'All, employed as his envoy to the Mongols in 658/1260.114 If we can trust

Wassaf, the future sovereign Jalal al-Din Khaljl was a refugee from Mongol territory in the 1260s who had

held command (amarat) of the Khalaj on behalf of the ruler of Binban;115 if he is the person referred to by

Juzjani as 'the Mongol "resident" (shihna) in Binban, who was the son of the amir Yughrush', Jalal al-Din

himself may have accompanied the Mongol embassy to Delhi in 658/ 1260.116 Barani, who lists him and his

brother Shihab al-Din Mas'ud among Balaban's maliks, has Jalal al-Din later recalling with emotion

Balaban's enthronement.117

Mongol immigrants, by contrast, were a new element in the politics of the Sultanate. In much the same way as Mongol commanders and their followers, worsted in some conflict with their confreres, began to seek

asylum in the dominions of the Mamltik Sultans of Egypt and Syria after 660/1262,118 so Mongol notables

fled into the territories of the sultan of Delhi. According to a fourteenth-century author, a whole quarter of

the capital was named 'Chingizi' after them in Balaban's era. If not already Muslims, they at any rate

embraced Islam after their arrival and, like their

112 Nasir al-Din Kirmani, Simt al-'Ula li'l-Hadratil-'Ulya (c.1315), ed. c Abbas Iqbal (Tehran, 1328

Sh./1949), 49. Anonymous, Ta'rlkh-i Slstan, ed. Malik al-Shu'ara Bahar (Tehran, 1314 Sh./1935), 405, and

tr. L. P. Smirnova (Moscow, 1974), 376. Hamd-Allah Mustawfi Qazwini, Ta'rikh-i Guz'ida, ed. 'Abd al-

Husayn Nawa'i (Tehran, 1339 Sh./1960), 532 (but with the wrong year, 669, for his flight). Rashid al-Din

says that Hajjaj remained in India for almost fifteen years: JT, II, ed. E. Blochet, GMS, XVIII (Leiden and London, 1911), 552, and tr. J. A. Boyle, The successors of Genghis Khan (London and New York, 1971),

305/tr. Iu. P. Verkhovskii, Sbornik letopisei, II (Moscow and Leningrad, 1960), 198. Brief references in

Wassaf, 291, and in Shabankara'I (738/1337), Majma' al-Ansab, ed. Mir Hashim Muhaddith (Tehran, 1363

Sh./1984), 199. That the Kirman ruler left Sistan on the news of the Ilkhan Abaqa's advance into Khurasan

places his departure in 678/1279: see JT, III, ed. A. A. Alizade and tr. A. K. Arends (Baku, 1957), text 152-

3, tr. 94, and for the date, Jean Aubin, 'L'ethnogenese des Qaraunas', Turcica 1 (1969), 85 n.4.

113 TN, II, 46 (tr. 798). TMS, 47. '14 TFS, 24. TN, II, 86 (tr. 860).

115 Wassaf, 311; JT, ed. Jahn, Indiengeschichte, Ar. text Taf. 57 (Pers. text Taf. 23 corrupt; cf. German tr.

48-9); QashanI, 185.

116 TN, II, 88 (tr. 862). TMS, 56, 61, calls his father Yughrush (Bras in the mss.).

117 TFS, 24, 178. For Shihab al-DIn, see also ibid, 186; DR, 54; Ibn Abi'l-Fada'il, ed. Kortantamer, Ar. text

29 (German tr. 108).

118 D. Ayalon, 'The Wafidiya in the Mamluk kingdom', IC 25 (1951), 81-104, repr. in his Studies.

~ successors who entered the Sultanate in the time of 'Ala' al-Din Khalji, are designated as 'neo-Muslims'

by Barani.119 Maliks with unmistakably Mongol names among the nobles of Balaban and his successor

include Bayanchar; Ulaghchi, the son of Turghai, Balaban's 'chief armour-bearer of the left hand' (sar-i

sildhdar-i maysara); Turumtai, one of the commanders who failed to suppress the Bengal revolt; and Ja'urchi, Kayqubad's sar-i jandar.120

These neo-Muslim amirs seem initially to have formed part of the coalition that secured the throne for

Kayqubad and to have shared power with Nizam al-Din, since they are described as enjoying office and

favour (shughldar-u muqarrab). But having disposed of other competitors, such as the wazir, Nizam al-Din

turned against them also. The Mongol amirs, including his former allies Kerei and the sar-i jdndar Ja'urchi,

were rounded up and most of them executed, although Ja'urchi and Mughaltai were merely exiled.121 Many

of Balaban's mawldzddas who were related to them by marriage were also eliminated, notably Malik

Shahik, entitled Azhdar Khan, who was amir-hajib and muqta'of Multan, and Malik *Turki, the 'arid122

It

looks very much as if Nizam al-Din, rather than Balaban, did most to undermine the Turkish ghulams during the few years prior to the seizure of power by the Khaljis.

Even after Nizam al-Din's removal, however, Mongol amirs were still at large. 'Isami may be correct in

naming those who later murdered Kayqubad as the sons of *Turki, i.e. Kayqubad's 'arid who had perished

during Nizam al-Din's ascendancy. That the ex-sultan is said to have been wrapped in his bedclothes and

kicked to death recalls pagan Mongol practice, which did not permit royal blood to be spilled on the

ground.123 And members of Balaban's slave establishment who had weathered the purge were able to

assume power: it was 'Balaban's slaves among the maliks, amirs, nobles and military commanders' who

despaired of the ailing Kayqubad early in 689/ 1290 and endeavoured to rule through his infant son

Kayumarth.124 Again we see a coalition of different elements, since this group, headed by Aytemur

*Kachhan and Aytemur Surkha, who on Nizam al-Din's downfall

119 Firishta, I, 131, citing Bijapuri's mulhaqat to TN. TFS, 133.

120 Bayanchar's name is garbled as NAHJN in TFS, 126, and as 'HHN ibid, 183 (for the correct form BAYNJR,

see BL ms, fols. 67b, 99a). Ja'urchi: TMS, 53, and FS, 184-8 (tr. 315-19). Turumtai: FS, 165-6 (tr. 292-3).

Turghai and Ulaghchi: TFS, 24, 126, 183. The father's name is actually Tu. turghai, iark', 'sparrow':

Sauvaget, 'Noms et surnoms', no. 128; Clauson, Etymological dictionary, 541. For Mo. ulaghchi, 'officer of

the postal relay-system', see Paul Pelliot, Notes sur Vhistoire de la Horde d'Or (Paris, 1950), 34-5; for

Bayanchar (Mo. bayan, 'rich', + suffix char), ibid., 52, 89; and for Mo. turumtai, 'male hawk', see F.D.

Lessing, A Mongolian-English dictionary (Berkeley and Los Angeles, 1960), 827.

121 TFS, 133-4. TMS, 53. FS, 186-8 (tr. 317-19). 122 TFS, 134. TMS, 55-6.

123 TFS, 173. Cf. J. A. Boyle, 'The death of the last 'Abbasid Caliph: a contemporary Muslim account', JSS

6 (1961), 150, repr. in his The Mongol world-empire (London, 1977).

124 TFS, 171.

~ had become respectively barbeg (amir-hajib) and wakil-i dar, allowed the Khalaj Jalal al-Din to be

summoned from his iqtac of Samana and given the office of 'arid and the,iqta' of Baran, with the style of

Shayista Khan.125 But the two Aytemurs sought to destroy Jalal al-Din. Aytemur *Kachhan fell in the

struggle, while Surkha was killed in a vain bid to secure Kayumarth.126

Jalal al-Din's rise to power appears to have been the product of a compromise. As Barani admits, a number of Turkish maliks and amirs had thrown in their lot with him,127 and there had been negotiations with the

Ghiyathi party, headed by Balaban's nephew Malik Chhajju; though why Chhajju refused Jalal al-Din's

offer of the dignity of na'ib and opted to retire to the iqta' of Kara is not explained. Jalal al-Din's own

enthronement in 689/1290 was the signal for Chhajju to revolt in Awadh at the head of Balaban's Turkish

slaves and their families and certain of the 'neo-Muslim' Mongol amirs. Chhajju and many of his supporters

were captured; Jalal al-DIn is said to have treated them leniently, though they forfeited their iqta's and

offices and Chhajju himself, who was sent to Multan, is not heard of again.128

The 'Khaljl Revolution'

Jalal al-Din's accession marked a break with the past in a way in which Balaban's usurpation had not. Early Muslim geographers and historians had regarded the Khalaj as a Turkish people;129 but accounts of the

transfer of power from the Ghiyathids to the Khaljis indicate that in late-thirteenth-century Delhi they were

regarded as a race quite distinct from the Turks.130 This may well be due to the particular sense acquired by

the word 'Turk', which in large measure had come to mean a Turkish ghulam (appendix I).

125 TMS, 55-6. For the title Shayista Khan, see TFS, 126, 170 (SYAST; but cf. BL ms., fol. 91b).

126 I have largely preferred the circumstantial account in TMS, 56-61, to that of TFS, 172-3, where these

events appear to be conflated to form a single episode, with *Kachchan's death closely followed by

Surkha's. The version in FS, 203-9 (tr. 365-72), seems to belong to the same tradition as Sirhindi's. On one point Sirhindi is definitely in error. He calls the child sultan Kayka'us (actually the name of Bughra Khan's

younger son and successor at Lakhnawti), but Barani's Kayumarth is corroborated by QS, 137, 142 etc., and

by Ibn Abi'l-Fada'il, ed. Kortantamer, Ar. text 29 (German tr. 108).

127 TFS, 172'.

128 The rebels included Balaban's mawlazada Amir 'Ali Hatim Khan, sar-i jandar and muqta' of Awadh;

two other former Ghiyathi commanders, Azhdar Khan's son Ikhtiyar al-Din Alp Ghazi, muqta' of Kasrak

(who was killed), and Malik Bahadur; and the 'neo-Muslim' amirs Bayanchar and Ulaghchi. HN, 313-15.

The sources are TMS, 62-4; TFS, 181-4; and (the most detailed) MF, 7-22. For Alp Ghazi as one of

Kayqubad's maliks, see TFS, 126 (omitting 'LP; but cf. BL ms., fol. 68a). His parentage is given in GK, IOL

Persian ms. 412, fol. 284a ('ZDR JAN in error for 'ZDR XAN); cf. also WH, ibid., fols. 114b, 143a (piir-i Azhdar Malik).

129 References are conveniently collected in Aziz Ahmad, 'Early Turkish nucleus', 103-5. See also

Shabankara'I, 87, who calls Jalal al-DIn 'likewise a Turk from among the Turkmen Khalaj' (ham turklbudaz

tarakima-yi khalaj).

171-2; cf. also 150, az asl-u qawm-i digar.

~ The transfer of power to Jalal al-Din was greatly resented by the notables of Delhi, members of the great

households (khaylkhanaha), many of whom may have been Turkish ghulams or their offspring and had

been ensconced in the capital since at least Balaban's day.131

Whether the change of dynasty had profound implications for the composition of the aristocracy, however,

by diluting the Turkishness of the governing class, is another question. Nobles of Turkish slave ancestry

may indeed have been the principal casualties of the Khalji seizure of power. Barani portrays the sons of

Balaban's maliks during Jalal al-Din's reign as a pool of dispossessed nobles, alert for opportunities to

undermine the new regime. For a time it seems that various grandees from Balaban's era who had lost their

offices and stipends attached themselves to the new sultan's eldest son, Mahmud, entitled Khan-i Khanan;

but on his premature death in c. 691/1292 they engaged in a conspiracy to replace Jalal al-Din with the

dervish Sidi Muwallih and to redistribute court offices and iqta's among the sons of Balaban's khans and

maliks. By no means all of them were Turks: they included the former grand qadi Kasani, *Hatya Paik and

the kotwal *Birinjin.132 Although the charges could not be proven, two of the accused nobles were executed

and the rest despoiled of their property and banished to outlying regions.133 The conspiracy seems to have been divulged to the sultan by a Mongol amir named Alughu, who had joined his court in 691/ 1292.134

As we might expect, the new sultan took care to promote fellow-Khalaj tribesmen, particularly members of

his somewhat large family, which comprised at the very least three sons; a brother, Malik Khamush, who

became 'arid; an uncle; and four nephews, one of whom was 'Ala' al-Din Muhammad (the future sultan),

the offspring of Jalal al-Din's deceased elder brother, Shihab al-Din Mas'ud. Other newly promoted Khalaj

amirs include the sultan's kinsman Ahmad-i Chap (at one time chamberlain to Aytemur Surkha, and so

called from the clipped pronunciation of hajib by the Khalaj), who became sar-i jandar-i maymana, and

probably Malik Iwad, who bore a name common among the Khalaj.135

Yet Sirhindi, who asserts that the majority of posts went to the new sultan's kinsfolk,136 overstates the case. An examination of the nobles listed

131 Ibid., 172, 173.

132 For the kotwal's name, which appears as BRNJTN in TFS, see BL ms., fol. 113b (BRNJYN). Birinjin had

presumably succeeded Fakhr al-Din, whose death during Jalal al-Din's reign is mentioned only by late

sources: AHG, II, 782; Firishta, I, 161. For Khan-i Khanan's personal name, see GK, cited in Mirza, Life

and works, 83 and n.3.

133 TFS 210-11. For differing interpretations of this episode, see Hodivala, Studies, I, 267-8, and Simon Digby, 'Qalandars and related groups', in Friedmann, Islam in Asia, I, 67-8.

134 TMS, 65; for his arrival, see TFS, 218-19.

135 TMS, 62. Jalal al-DIn's relatives are listed in FS, 226-7 (tr. 392-3). On Ahmad-i Chap, see Hodivala,

Studies, I, 266.

136 TMS, 62.

~ by both BaranI and Sirhindl in their accounts of Jalal al-Din's reign reveals that a large proportion had

been prominent under the Ghiyathids and did not belong to the Khalaj. The most obvious instances are

Khwaja Khatir al-Din, who had been disgraced by Nizam al-Din but who was now restored as wazir, and the kotwal Fakhr al-Din, who was confirmed in office.137 Malik Fakhr al-Din Kuchi and his brother Malik

Taj al-Din possibly belonged to a family which had produced amirs in Iltutmish's day; under Jalal al-Din

they are found acting respectively as dadbeg and as muqta' of Awadh.138 Other examples of nobles who had

served the Ghiyathids are the Indian converts Malik cAyn al-DIn 'All Shah KuhijudI and his brother Malik

Ikhtiyar al-DIn Khurram, of whom the latter was now promoted to wakil-i dar.139 Such examples could be

multiplied.140 We even find Ikhtiyar al-Din-i Hindu Khan-i Ghiyathi - from his cognomen (nisba) evidently

the son of one of Balaban's slaves - as nd'ib-i wakll-i dar under the new sultan.141 Although some of these

men would later be implicated in a half-hearted conspiracy against Jalal al-Din,142 it is noteworthy that

none of them is said to have supported Chhajju's revolt and that the KuhijudI brothers and Taj al-DIn Kuchi

fought under the Khalji banners on that occasion.143

The opening years of the Khaljl dynasty exhibit a striking continuity with the preceding era; and the ruling

elite following Jalal al-Din's accession bears the stamp of compromise that we noticed in connection with

earlier changes of regime. By contrast, the impression given is that a dramatic shift in the composition of

the ruling class - the real 'Khaljl revolution' - came

137 TFS, 174, 177.

138 For Taj al-Din and Fakhr al-Din, see also ibid., 203; MF, 14. Their father may have been Balaban's

dadbeg, Malik Nasir al-Din Kuchi: TFS, 24 (though FS gives this as the name of Fakhr al-Din's brother).

For earlier Kuchi maliks, TN, I, 456, 458, 459, and II, 6, 13 (tr. 633-4, 639, 640, 726, 735).

139 TMS, 84 (KHJWRY in error). MF, 27, 33-4. TFS, 174 ('XBAR for 'XTYAR, to be corrected from BLras, fol. 93b), 177, 195, 233.

140 Malik Ikhtiyar al-Din Begtiit, nd'ib-i amir-hdjib under both Kayumarth and Jalal al-Din: TMS, 60, and

TFS, 126 (SKNT, to be corrected from BL ms., fol. 68a); and for the name, L. Rasonyi, 'Les noms de

personnes imperatifs chez les peuples turcs', AOH 15 (1962), 241-2. Malik Mughaltai, previously exiled by

Nizam al-Din: TMS, 64-5. Malik Naslr al-Din Rana, who retained his office of shihna-yi pil from Balaban's

reign through to that of 'Ala' al-Din Khaljl: TMS, 58. Malik Nusrat-i sabah, whose father had likewise been

a malik and who now became chief inkwell-holder (sar-i dawdtdar): TFS, 174 (JNAH for SBAH, to be

corrected from BL ms., fol. 93b), 198, 204; FS, 227 (tr. 393); MF, 14. Others mentioned in MF include

Malik 'Ayn al-Din *Hiranmar, who became Jalal al-Din's amlr-i shikar; Malik Mahmud, Jalal al-Din's sar-i

jdndar; Malik *Kiki, who governed Kol for the new sultan. 141 TMS, 62, 69.

142 *Hiranmar forfeited his office of sar-i jdndar and Taj al-Din seems to have been deprived of Awadh,

but Mughaltai and Mubarak were merely banished to their iqta's for one year: TFS, 190-2; TMS, 64-5.

143 Except where other references have been given, the reconstruction in this paragraph is based on TFS,

126, 174, 177; TMS, 54, 62; and MF, 14, 28. Nigam, Nobility, 53 and Appendix C, reaches conclusions

broadly similar to mine, though relying on the corrupt readings in the printed text of TFS.

~ with 'Ala' al-Din. Initially the new sovereign was careful to reward those Jalali grandees who

had deserted his cousins. But once 'Ala' al-Din's henchmen had secured the surrender of Multan in

Muharram 696/ November 1296 and Jalal al-Din's sons and their supporters had been blinded, the regime

was strong enough to move against the Jalali amirs. In the second year of the reign (late 696-late 697/1297-

8), the great majority of them, including Hiranmar (now Amin Khan), who had briefly held Multan, and

Abachi (Arslan Khan), were arrested and imprisoned, blinded or executed; their wealth, iqta's and offices

were confiscated, and their military contingents transferred to 'Ala' al-Din's own clientela. According to

Barani, only three, who had not betrayed Jalal al-DIn's sons or accepted gifts from the new regime, were

spared.144 Even this did not end the purge of the old nobility. The Malik al-Umara' Fakhr al-DIn, kotwal of

Delhi, had probably died during Jalal al-DIn's reign; but his sons were rounded up and executed in

700/1301 on suspicion of complicity with a rising in Delhi led by their freedman Hajji Mawla.145 Barani

exaggerates when he claims that in his own day no descendants of Balaban's nobles survived (see below, pp. 189-90);146 but there can be little doubt that their almost total disappearance dates from the reign of'

'Ala' al-Din.

144 TFS 242, 247-8, 249-51. TMS, 71-2.

145 TFS, 282. For Fakhr al-DIn's death, see above, n . 132.

146 TFS, 48; cf. also HN,302.

~CHAPTER 5

The centre and the provinces

The terminology applied to the sultan's dominions was frequently unspecific. Juzjani writes of 'the

empire of Delhi' (mamalik-i Dihli) and Barani of 'the provinces of the Delhi empire' (bilad-i mamalik-i

Dihli).1 Ibn Battuta speaks of Muhammad b. Tughluq's empire as 'Hind and Sind',2 distinguishing the

territory that had been won for Islam in the eighth century from the rest of the subcontinent. A Muslim

geographer of an earlier generation distinguishes 'Hindustan', the conquests of the Ghurids and their

epigoni, from the wider Indian world, which he terms 'Hind'.3 This usage echoes that of writers within

India. JuzjanI sometimes calls the Sultanate, like the Ghaznawid and Ghurid conquests before it, 'the terri-

tories (mamalik) of Hindustan';4 and we read accordingly of rulers obtaining the 'throne' (takht), or 'the

kingdom' (mulk), of Hindustan.5 But in the thirteenth and fourteenth centuries, confusingly, 'Hindustan' also

had a narrower significance for Muslim authors within the subcontinent. It denoted the Doab - the mesopotamia (miydn-i du db) between the Yamuna and the Ganges - along with the other partially

subjugated regions to the east and south-east. People flee from Delhi 'to Hindustan';6 we hear of 'the iqta's

in the direction of Hindustan' (simt-i Hindustan)7 and of the amirs and troops of Hindustan, clearly in this

restricted sense of the regions of the Doab, Awadh and Kara.8 Such latitude in the use of geographical

terms can

1 TN, II, 39 (tr. 785). TFS, 468.

2 IB, III, 94, 161, 215, 248 (tr. Gibb, 593, 628, 657, 674); see also al-Safadi, Waft, III, 172.

3 Ibn Sa’id al-Maghribi, Kitab al-Jughrafiyya, 134, 163. 4 TN, I, 6, 398, and II, 88, 90 (tr. xxxii, 455, 863, 874-9); cf. also I, 418 (tr. 530), where the term is

used of the whole of Muslim-ruled territory in the subcontinent, including Sind; also II, 9, 32, mamlikat-i

Hindustan (tr. 731, 764).

5 Ibid., II, 162, 169 (tr. 1129, 1153). TFS, 249. FS, 604, 605 (tr. 898, 899).

6 TN, II, 49 (tr. 802).

7 Ibid., II, 66 (tr. 830): they comprised 'Kara and Manikpur, Awadh and Tirhut as far as Bada'un'.

TFS, 272, 300. TMS, 63.

8 TN, I, 453, and II, 15, 29, 66 (tr. 629, 739, 759, 760, 830). The person in command was usually

the muqta' of Awadh. Ibid., II, 72 (tr. 839), 'Hindustan' clearly denotes Awadh, as distinct from the hill-

state of Santur; but cf. II, 58-9 (tr. 818), where it also implicitly includes the Chawhan kingdom of

Ranthanbor. TFS, 57, 141, 181, 182, 257, 300, 301, 328.

86

~ give rise to confusion. Barani strikes perhaps the most incongruous note of all when he describes

Sultan Balaban's vanquished 'Hindustani' troops (from Awadh) being despoiled by 'Hindus'.9

Even in the thirteenth century the Sultanate technically embraced a vast area and comprised

several regions which could be termed kingdoms in their own right. Lahore, as 'the residence of Khusraw

Malik'10 (the last Ghaznawid sultan) and successively the appanage of Iltutmish's two elder sons (pp. 30, 46

above), was one. Another was Lakhnawti, wrested by Iltutmish with such difficulty from its Khalaj rulers

and extensive enough to be termed a 'clime' (iqtim).11 Both Juzjani and BaranI speak of the 'throne' (takht)

or 'kingship' (mulk) and the 'insignia of rulership' (padishahi) being conferred on those sent to govern

Lakhnawti - whether Balaban's son in c. 680/1281 or great amirs like 'Ala' al-Din Jani and (later) his son

Qilich Khan Mas'ud.12 Viceroys of this eminence received certain quasi-imperial privileges such as the

right to the durbash, or baton, and the chatr or ceremonial parasol: thus Iltutmish sent his son Nasir al-DIn

Mahmud in Bengal a red chatr in 626/1229, and Balaban conferred a chatr and a durbash on one son,

Bughra Khan, at Lakhnawti, and a chatr on the other, Muhammad, at Multan, later transferring this emblem of authority to Muhammad's son Kaykhusraw when his father died.13 By this time Multan had replaced

Lahore as the territory allotted to the heir-apparent, though whether Jalal al-Din Khaljl, in conferring the

city on his son Erkli Khan, granted him the customary insignia we are not told. The policy seems to have

continued into the fourteenth century, when 'Ala' al-Din Khaljl bestowed similar insignia - a red chatr, a

robe of honour, and two standards - on his eldest son Khidr Khan at Chitor in 703/1303, in much the same

way as he conferred a chatr on the submissive Yadava king of Deogir in 706/1307.14

The Delhi Sultanate could not, perhaps, be clearly defined in spatial terms. During the thirteenth

century it should be seen as a collection of sub-kingdoms,15 some ruled by Hindu potentates who

periodically rendered tribute, others by princes of the sultan's dynasty or by Muslim amirs and muqta's.16

What ultimately determined the extent of the monarch's rule was recognition by the provincial governors,

particularly those of outlying regions. Juzjani s claim that at Radiyya's accession 'all the maliks and amirs from the territory of Lakhnawti as far as Diwal (Daybul) and Damrila

9 Ibid., 84.

10 TN, I, 454 (tr. 631).

11 E.g. TFS, 82, 93.

12 Ibid., 82, 92. TN, I, 448, and II, 13, 31, 35, 78.

13 Ibid., I, 454 (tr. 630). TFS, 66, 92, 110; for Bughra Khan, see also DGK, 69. 14 Yadava king: KF, 63-4; TFS, 326. Khidr Khan: DR, 67; TFS, 367; TMS, 11.

15

A point made by Peter Hardy, 'The growth of authority over a conquered political elite: the early

Delhi Sultanate as a possible case study', in John S. Richards (ed.), Kingship and authority in South Asia

(Madison, Wisconsin, 1978), 203-4.

16 This is implicit in the views which BaranI puts into Balaban's mouth: TFS, 93.

~ manifested submission',17 is (whether true or not) a more idiomatic statement concerning a sultan's authority than any amount of grandiloquence about the throne of 'the whole of the empire of

Hindustan'.

Disintegration and recovery

During the thirteenth century the empire created by Iltutmish fluctuated considerably in extent, as

the conflicts within the ranks of the aristocracy, particularly the struggle between Ulugh Khan Balaban and

his enemies in the 1250s, were often played out also in the provinces. That we know less about

developments in territories at a distance from Delhi reflects in some measure the nature of the source

material. The regional histories spawned from the fifteenth century onwards by the successor states fail to

supplement Juzjani's testimony for the thirteenth, and apart from the period of his exile in Lakhnawti (640-

3/1242-6) Juzjani's perspective is always that of the centre and the court. Nevertheless, his tabaqa 22 - on the Shamsi maliks -does furnish a good deal of data on the iqta's.

Following Iltutmish's death in 633/1236, Juzjani tells us, respect for 'the kingdom of Hindustan'

suffered a sharp decline, so that rivals sprang up on all sides and desired to appropriate its territories.18

Many of these unspecified enemies would have been Hindu princes, while in Sind the most formidable

adversary confronting Iltutmish's successors was initially the former Khwarazmian lieutenant Hasan

Qarluq. But the sultan's own officers also profited from the situation at the centre. Juzjani's cryptic remark

that the Shamsi slave Sayf al-Din Aybeg, muqta' of Uchch, 'grew powerful' on Iltutmish's death might

suggest that he briefly asserted his independence of Delhi before he was killed in a riding accident.19 Sind

is expressly included among the territories that submitted to Radiyya, and the next officer in charge of

Uchch of whom we are told was her appointee, Hindu Khan, who was removed after her deposition.20 The Mongol invasion of the Sultanate early in 639/1241-2 furnished new opportunities for self-aggrandisement.

At Lahore Qaraqush Khan Ikhtiyar al-Din Aytegin, who had been installed there by Radiyya after the

suppression of Kabir Khan's revolt in 637/1239-40 (p. 67 above), had supported her attempt to regain the

throne in the following year,21 and so was at this time technically in rebellion against Bahrain Shah. An

army sent from Delhi turned back to besiege the capital and overthrow the sultan; but its purpose - not to

relieve Lahore, we are told, but merely to guard the frontier22 -

17 TN, I, 459, omitting Damrila; but cf. BL ms., fol. 182b (also Raverty's tr., 641).

18 Ibid., II, 9(tr. 730-1).

19 Ibid., II, 8-9 (tr. 730-1): the phrase qui-yi hdl gashta is omitted in the printed text, but is found

in BL ms., fol. 199a.

20 TN, I, 459 (tr. 641); II, 19 (tr. 746), for Hindu Khan.

21 Ibid., I, 462 (tr. 647).

22 Ibid., I, 466 (tr. 657).

~ reflects the fact that under Qaraqush Khan the city was now regarded as lying outside the

sultan's dominions. In Multan Kabir Khan Ayaz had proclaimed his independence with the adoption of a

chatr, and occupied Uchch.23 At this critical juncture, therefore, neither Lahore nor Sind formed part of the Sultanate.

Kabir Khan Ayaz and his son Taj al-DIn Abu Bakr in turn defended their principality against

Hasan Qarluq. The rule of this shortlived dynasty, for which Professor Nizami coined the appropriate name

of 'Ayazi', ended with Abu Bakr's death in the early 1240s, when Hasan Qarluq finally obtained Multan.24

Although Sind was recovered by the sultan's forces in the wake of the Mongol invasion of 643/1245,25 the

risk of secession grew with the onset of the conflicts of Nasir al-Din Mahmud Shah's reign. Sind effectively

ceased to form part of the Sultanate shortly afterwards, when Kushlu Khan, thwarted in his joint bid with

Qutlugh Khan to seize Delhi (pp. 74-5 above), appealed to the Mongols. Juzjani, writing when the Mongols

had dismantled Multan's fortifications, betrays by his phrasing that Sind now lay outside 'the borders of the empire of Delhi' (sarhaddhd-yi mamalik-i Dihli).26

Lahore is described as ruined (khardb) in the wake of the Mongol sack in Jumada II

639/December 1241,27 although the city seems to have been held for a couple of years by Ikhtiyar al-Din

Yiizbeg (Toghril Khan).28 There after Juzjani makes little mention of Lahore,29 which like the regions lying

beyond it, in the far north-west, had apparently come to form part of the Mongol dominions. From c.

651/1253 we find the renegade prince Jalal al- Din Mas'ud b. Iltutmish at Lahore as a Mongol client (p. Ill

below). He maintained himself there only for a short time, being dislodged by Shir Khan on the latter's

return from the Mongol court.30 Although he once more became an ally of the Delhi government, Shir

Khan soon began to evince designs on his old iqta' of Tabarhindh, which was currently held by Arslan

Khan, and engaged in conflict with him too. At this point the court interposed, winning Shir Khan over

with the grant of Tabarhindh and 'the whole of the territory and iqta's which he had previously held', including presumably Lahore. Yet the place was doubtless impossible to hold, and Shir Khan seemingly

abandoned it. When in 657/1259 the sultan obliged

23 Ibid., II, 6 (tr. 727).

24 At Uchch Hindu Khan's officials, however, remained in place: for this and Hasan Qarluq's

seizure of the city, ibid., II, 169-70 (tr. 1153). 'Ayazi dynasty': HN, 249, 255, 260. On Abu Bakr, see

Siddiqi, 'Historical information', 64.

25 TN, II, 28, 37 (tr. 758, 781).

26 Ibid., II, 39 (tr. 785); and cf. also II, 86 (tr. 860).

27 Ibid., II, 169 (tr. 1153). For the date, see below, p. 105 and n.12.

28 TN, II, 30 (tr. 762).

29 Ibid., II, 43, has Lahore being granted to Shir Khan along with Tabarhindh in 643/1245, but cf.

BL ms., fol. 211 (also Raverty's tr., 793), where Lahore is omitted.

30 TN, II, 44 (tr. 793).

~ him to exchange his extensive assignment with Nusrat Khan, the muqta' of Bhayana, there is no mention of Lahore among the latter's new holdings.31

To the east, Awadh could be characterized as another problematic region in the 1250s. Qutlugh

Khan, who was sent there after his removal from court in 653/1255, encroached on the territory of Bada'un,

and defeated a force under its muqta' *Teniz Khan, reinforced by troops from Delhi under Begtemur Or

Khan-i Rukni; Or Khan was killed.32 In 654/1256 Awadh was entrusted to Arslan Khan, who had been a

prominent supporter of Ulugh Khan Balaban two years before; and the new muqta' performed sterling

service by obstructing Qutlugh Khan's efforts to occupy Kara. But then, says Juzjani, Arslan Khan's attitude

towards the government underwent a change and he grew rebellious. When preparations were under way at

Delhi early in 656/1258 for a campaign to drive the Mongols from Sind, Arslan Khan and Qilich Khan Jalal

al-Din Mas'ud, the son of 'Ala' al-Din Jani and muqta' of Kara, neglected to bring their contingents. Ulugh Khan Balaban marched on Kara, and the two recalcitrant amirs were brought to heel.33 Qilich Khan

received a patent for Lakhnawti; Arslan Khan was transferred to Kara, although as we shall see his new

iqta' was not of a size to contain his ambitions.

Beyond Awadh, tenuous links bound Delhi to the distant Muslim-held territories in Bengal. In the

fourteenth century this region would be known as the 'lowlands' (furu-dast),24 to distinguish it from the

'upper country' (bald-dast), which vaguely embraced the north-west and the lands beyond the Indus

towards Transoxiana. 'Afif, a westerner of course, makes the 'lowlanders' (furudastan) confess to being no

match for those from the uplands (baladastdn)?5 Yet to conquer and hold Bengal - 'a land for foot-soldiers'

(zamln-i rijdla), as the Tughluqid Sultan Firuz Shah would describe it in 135436 - was no easy task. It was proverbially wealthy - 'a Hell full of dainties' (duzakh ast pur-i nimat), to quote Ibn Battuta.37 None of the

amirs of Hindustan, says Barani, could rival in terms of men, elephants or treasure whoever controlled

Bengal; and the conduct of many of its governors during the thirteenth century fully vindicated the

nickname 'Bulghakpur' ('city of insurrection') which he says men gave to Lakhnawti.38

After Iltutmish's death, the sultans had to recognize the autonomy of

31 Ibid., II, 42-3 (tr. 788); for the exchange, see also II, 44 (tr. 794).

32 Ibid., I, 490, and II, 29 (tr. 703, 759-60).

33 Ibid., II, 34-5, 77-8 (tr. 768-9, 847-8). For Qilich Khan, who has frequently been confused with

other grandees, see appendix II.

34 TFS, 189. SFS, 33, 48. Bihamadkhani, fol. 421b, uses the term more broadly, for the regions

east of the Yamuna.

35 'Afif, 153; for the sense, see Hodivala, Studies, II, 127.

36 'Afif, 119. On this phrase, see Hodivala, Studies, I, 312-13, who translates it as 'a land of foot-

soldiers': in view of the phrase that follows regarding the difficulty of life among the islands, I suspect that

the sultan meant, rather, 'a land [fit only] for foot-soldiers', i.e. that the heavily armoured horsemen of the

Delhi Sultanate were ineffective here. 37 IB, IV, 210 (tr. Gibb and Beckingham, 867).

38 TFS, 82.

~ their representatives at Lakhnawti while receiving little in return. Since 631/ 1233-4 the muqta'

of both Bihar and Lakhnawti had been the Shamsi slave 'Izz al-Din Toghril Toghan Khan. The

geographical location of his iqta' conferred on Toghan Khan a good deal of independence, and once

Iltutmish was dead he engaged with impunity, it seems, in warfare against the muqta' of Lakhnor (the

region on the west bank of the Ganges), slaying him and appropriating part of his territory. Both Radiyya

and Bahram Shah nevertheless legitimized Toghan Khan's irregular actions by sending him a red chatr and

standards; and after the accession of Mas'ud Shah, Toghan Khan dropped all pretence of loyalty to the Delhi court. An inscription from Bihar dated 640/1242 accords him a variety of grandiose titles and makes

no reference to the sultan. In this same year he advanced into the Kara-Manikpur region in an abortive

attempt to occupy Awadh.39

Following a disastrous campaign against the Hindu kingdom of Jajnagar (Orissa) late in 641/early

in 1244, however, Toghan Khan asked the sultan for reinforcements. Mas'ud Shah thereupon despatched to

Lakhnawti not merely a red chatr and a standard but a robe of honour and an ornate tent. But these

attentions were apparently designed simply to throw Toghan Khan off guard. We learn elsewhere that Taj

al-Din Sanjar *Kirit Khan died from an arrow-wound outside the city of Bihar in an obscure conflict soon

after 640/1242,40 which - unless the place had been lost by the Muslims and the enemy was a Hindu king -

suggests that the government may already have been attempting to retrieve Bihar. Now, on Toghan Khan's appeal for help, orders were issued to the muqta' of Awadh, Temur Khan, and other commanders to move

into Bengal. When this supposedly relieving force reached Lakhnawti, it engaged in hostilities not with the

Hindus, who had withdrawn, but with Toghan Khan. Compelled to come to terms, he surrendered the

province to Temur Khan and accompanied the other amirs back to Delhi. At this juncture Mas'ud Shah's

government indulged in another attempt to play the two enemies off against each other, granting Toghan

Khan in 643/1245 Temiir Khan's iqta' of Awadh. Temiir Khan's reaction to the seizure of Awadh from his

officers is not described; but he remained in control of Lakhnawti until both men died, on the same day in

Shawwal 644/March 1247, during the reign of Nasir al-Din Mahmud Shah.

The history of Lakhnawti over the next few years is obscure. An inscription of 647/1249 commemorates as the current muqta' Jalal al-DIn Mas'ud (the later Qilich Khan), son of cAla' al-Din Jam.41

But he is not mentioned at this stage by JuzjanI, who next names Ikhtiyar al-Din Yiizbeg

39 It was on the borders of Kara that JuzjanI waited on him before accompanying him back to

Lakhnawti: TN, I, 469 (tr. 662-3). For Toghan Khan's career, see ibid., II, 13-17 (tr. 736-42). Yazdani,

'Inscriptions of the Turk Sultans', 16-17; RCEA, XI, 143 (no. 4215).

40 TN, II, 27-8 (tr. 757).

41 Yazdani, 'Inscriptions of the Turk Sultans', 19-22; RCEA, XI, 211 (no. 4320).

~ Toghril Khan in command of the province. The career of this compulsive rebel epitomizes the problems posed for the government by over-mighty subjects. On the accession of Mas'ud Shah, Yuzbeg had

been entrusted with Tabarhindh. Then - most probably at the time of that monarch's expedition to Uchch in

643/1245-6 - he was transferred to Lahore, where he first engaged in conflict with an otherwise unknown

malik and then defied the sultan's authority. As we have noted, Ulugh Khan Balaban secured for him both a

pardon and a new iqta' at Qinnawj, where he seems to have again proved refractory, probably in support of

Balaban. Reduced to obedience by an expedition under Qutb al-Din Hasan b. 'Ali, he was brought back to

court and subsequently assigned Awadh. It is likely that he supported Ulugh Khan Balaban's return to

power in 652/1254. From this date, however, Yiizbeg Toghril Khan is found in command at Lakhnawti.

Here a major victory over the raja of Jajnagar, who had successfully resisted Toghan Khan's aggression

over ten years previously, encouraged him to assert his independence of Delhi; he assumed three chatrs and

had coins struck and the khutba read in his own name as Sultan Mughith al-Din. He died a few years later, in the course of a disastrous invasion of neighbouring Kamrup (Assam).42

Yiizbeg Toghril Khan's death seems to have occurred before Safar 655/ February-March 1257,

when coins were again being struck at Lakhnawti in the name of Nasir al-DIn Mahmud Shah.43 At the end

of 656/1258 Lakhnawti was conferred once more on Qilich Khan Jalal al-DIn Mas'ud-i Jani, who had

recently been guilty of insubordination along with Arslan Khan. But only a few months later, in Jumada II

657/June 1259, Ulugh Khan Balaban persuaded the sultan to recognize as muqta' 'Izz al-DIn Balaban-i

Yiizbegi, whom we have already encountered as the son-in-law of the rebel Qutlugh Khan and who had

just secured the court's favour by sending impressive gifts to Delhi.44 In the event, however, Balaban-i

Yiizbegl lost the province not to Qilich Khan (who is not heard of again) but to the latter's confederate.

Acting without sanction from Delhi and giving out even to his own sons and troops that he was engaged in

a plundering expedition into infidel Malwa, Arslan Khan left Kara and marched swiftly on Lakhnawti. The city was taken and sacked, and Balaban-i Yiizbegl, who returned from operations in eastern Bengal (Bang)

to offer battle to the invaders, was captured and put to death. This was the state of affairs in the east when

Juzjani stopped writing in 658/1260.45

42 TN, II, 30-3 (tr. 762-6), for Toghril Khan's biography. His career is conveniently summarized by

Habibullah, Foundation, 129-30. For his coins, see now Roma Niyogi, 'A new coin of Mughisu-d-Din

Yuzbak of Bengal', JNSI40 (1978-9), 136-8.

43 CMSD, 55 (no. 225C). M TN, I, 495, and II, 35, 78 (tr. 712, 770, 848-9).

45 Ibid., II, 35 (tr. 769-72). Laiq Ahmad, 'Kara, a medieval Indian city', IC 55 (1981), 85, confuses

the events of these years and assumes that Arslan Khan's sack of Lakhnawti preceded the recalcitrance

mentioned above.

~ Barani's testimony may indicate that Arslan Khan remained defiant in Lakhnawti until his death,

since the despatch of elephants from Bengal to Delhi upon Balaban's accession in 664/1266 by his son and

successor, Tatar Khan Muhammad, was viewed as something out of the ordinary.46

Events in these outlying regions highlight the restrictions on the power of the sovereign. Amirs

frequently revolted; in distant Bengal, at least, they might do so with impunity for some years. In order to

assert itself, the government on more than one occasion resorted to acts of duplicity. The sultan might confer a rebel's territory on a rival only to make use of the dispossessed amir as a counterweight to the now

over-powerful commander who had taken his place, as seems to have transpired in 642/1244 first in the

removal of Toghan Khan from Lakhnawtl and then in his despatch to Awadh. Sometimes the court made

virtually simultaneous grants of the same iqta' to two different nobles and left them to fight it out. This

occurred in 653/1255, when the conferment of Bahraich on Taj al-Din Sanjar Siwistani followed

suspiciously close on the heels of its allocation to 'Imad al-Din Rayhan (p. 74). It may have been the policy

regarding Lakhnawti in 656/1258, when the regime's contradictory actions defy interpretation. Perhaps the

court had briefly envisaged the removal of Balaban-i Yiizbegi, as a surviving connection of Qutlugh Khan.

But it is equally possible that the reconciliation between Qilich Khan and the regime at Delhi was merely

superficial and that the grant of Lakhnawtl represented an attempt to double-cross him.

In the latter part of Mahmud Shah's reign the Delhi empire had contracted to the point where it embraced an area hardly larger than that ruled at his accession by Iltutmish. When Juzjani wrote, it might

well have appeared to be in a state of disintegration. Why it did not disintegrate, we shall probably never

fully understand. As regards the western provinces, a partial explanation is surely furnished by divisions

within the Mongol ranks (see below, pp. 108-10, 115-16). But such information as we have concerning the

Sultanate's internal history furnishes no answer. Between Juzjani's completion of his Tabaqat in 658/1260

and the commencement of Barani's history with the accession of Balaban stands a hiatus of six years that

may well have been crucial for the Sultanate's survival.

To say that the government's authority over its distant provinces was restored during Balaban's

own reign is to beg a large question, for the treatment of events by Barani, 'Isami and Sirhindi is far less

detailed than Juzjani's handling of the vicissitudes of Mahmud Shah's era; it cannot be emphasized sufficiently that we know more about Balaban as khan and na'ib than we do of his time as sultan.

Nevertheless, there can be no doubt that under Balaban some kind of rassemblement occurred of the

territories

46 TFS 53; see also 66, where he is called Muhammad and described as 'padishah of Lakhnawti.

~ that had at one time acknowledged Iltutmish. He is known to have led an army to Lahore within

a few years of his accession and to have restored and repopulated the city,47 so that it can once more be

deemed to have formed part of the Sultanate; though it is surely significant that we find no reference to its

being granted as iqta' again before the fourteenth century. Sind was also recovered. At some point towards

the end of Mahmud Shah's reign, Kushlu Khan lost control of the province, in obscure circumstances. In a

problematic passage, 'IsamI says that Ulugh Khan Balaban profited from Kiishlu Khan's absence to seize Multan.48 His disappearance may be connected with the advent of the Neguderi Mongols, which will be

examined later (see pp. 115-16). At any rate Balaban was able to instal as viceroy in Sind his own elder son

Muhammad (Khan-i Shahid),49 who governed until his untimely death in battle with the Mongols at the

very end of683/in March 1285.

Lakhnawti appears to have been regained with comparative ease, since Tatar Khan died in, or soon

after, 665/1266—750 and Balaban despatched his own representatives to the province. The sources differ,

Barani and 'IsamI alleging that the sultan's ghulam Toghril was sent out to Lakhnawtl as governor (wall),

while Sirhindi has Balaban appointing Amin Khan (i.e. Aytegin-i Mui-yi Daraz; see p. 78) to the post, with

Toghril as his deputy (na'ib).51 In any event, the province proved no less turbulent for Balaban than for his

Shamsid predecessors. Toghril rebelled in c. 678/1279-80, and in a gesture strikingly remininscent of Yuzbeg a generation earlier-assumed the style of Sultan Mughith al-Din. He defied two successive

campaigns by the sultan's lieutenants before he was overwhelmed by an army under Balaban in person,

probably in 680/1281-2.52

Balaban then entrusted Lakhnawtl to his younger son, Bughra Khan Mahmud, who proclaimed his

own sovereignty after the old sultan's death in 685/1287, when he found his expectations of the throne

cheated by his son Kayqubad, and briefly occupied Awadh.53 The status of Lakhnawtl after father and son

were reconciled in 686/1287 is therefore unclear; but

47 Ibid., 61. TMS, 40. Cf. also FS, 164 (tr. 291). 48 Ibid., 154-5 (tr. 278-80).

49 TFS, 66, dating this after the death of Shir Khan, which at 64-5 is said to have occurred 'four or

five years' into the reign.

50 For an inscription of Tatar Khan dated 665/1266-7, see Yazdani, 'Inscriptions of the Turk

Sultans', 23-5; RCEA, XII, 121-2 (no. 4854).

51 TFS, 81. FS,165 (tr. 292). TMS, 40.

52 TFS, 81, 83-92; at 81, Barani places Toghril's revolt 'fifteen or sixteen years' after Balaban's

accession, i.e. in 677-8/1278-80 if we start from his incorrect year 662 for the sultan's enthronement. This

can be reconciled with the report on Balaban's campaign against Jajnagar (Orissa) found in RI, V, 5-13, and

dated 5 Shawwal 680/17 January 1282 (ibid., 13). Firishta, I, 138, also dates the revolt precisely to 678,

though on what authority is unclear. FS, 164 (tr. 291), puts the rising a mere eight years after Balaban's

accession, and later, 168 (tr. 296), supplies the impossible year 670.

53 QS, 36, 44-6.

~ following the transfer of power in Delhi to Jalal al-Din Khalji in 689/1290 Bughra Khan and his

successors certainly acted as independent monarchs. Epigraphical evidence shows that their authority

extended also over the Muslim territory in southern Bihar.54 This branch of the Ghiyathid dynasty was shortlived. Bughra Khan's young son and successor, Rukn al-Din Kayka'us, who died around the turn of the

century, was followed first by an obscure ruler named Shams al-Din Dawlat Shah and then by Shams al-

Din Firuz Shah, who is probably identical with Kayka'us's amir Firuz-i Aytegin and who founded a new

dynasty.55 At some point 'Ala' al-Din Khalji may have attacked Bengal;56 but it was only with the outbreak

of a struggle among Firuz Shah's sons, leading to the intervention of Ghiyath al-Din Tughluq Shah in

724/1324, that the region again became a province of the Delhi Sultanate (see below, pp. 200-1).

The iqtai'and provincial government

At the heart of the thirteenth-century Sultanate lay the khalisa or 'reserved' lands - what might be

called the 'royal demesne' - from which the sultan's own officials collected revenue directly and which

provided his most immediate resources. Regarding the full extent of the khalisa we have no information, although it is usually taken to have included the environs (hawali) of Delhi.57 Other territories were granted

out as iqta'. The term 'iqta' applied not only to the large assignments enjoyed by great amirs but also to the

smaller ones established by Iltutmish in the Doab, according to Barani, who tells us that each grantee

(iqta'dar) was expected to raise from one to three horsemen.58 It is possible that Amir Khusraw's father,

Sayf-i Shamsi (d. c. 659/1261), held such an iqta'.59[Early in his reign Sultan Balaban sought to resume

many of these small iqta's into the khalisa, on the grounds that the grantees were now too old to serve or

had died and had transmitted their holdings to heirs who performed no service; in the event, says BaranI, he

was dissuaded by the kotwal Fakhr al-DIn.60 Although the chronicler does not say so, the episode was

doubtless a measure of

54 A. A. Kadiri, 'Inscriptions of the Sultans of Bengal from Bihar', EIAPS (1961), 35-6; Q. Ahmad,

Corpus of Arabic and Persian inscriptions of Bihar (AH 640-1200) (Patna, 1973), 9-10 (no. 3).

55 On Dawlat Shah, see John S. Deyell, 'A reassessment of the new coin of Daulat Shah of Bengal',

JNSI 41 (1979), 82-90. For Firuz Shah's line, see Abdul Majed Khan, 'The historicity of Ibn Batuta re

Shamsuddin Firuz Shah, the so-called Balbani king of Bengal', IHQ 18 (1942), 65-70; R. C. Majumdar,

History of mediaeval Bengal (Calcutta, 1973), 18.

56 According to the anonymous translator of Bahr al-Hayat, IOL Persian ms. 432 (Ethe, no. 2002).

TFS, 227-9, 254, speaks only of his designs on Bengal. An ode in GK (tr. in ED, III, 543) refers to a

campaign in Bihar and the seizure of Bengal elephants from Lakhnawti. 57 On the khalisa, see W. H. Moreland, The agrarian system of Moslem India (Cambridge, 1929),

29 and n.l; more generally, A. K. S. Lambton, 'Khalisa', Enc.Isl2

58 TFS, 62.

59 On him, see DGK, 66, 67: he died when Khusraw was seven.

60 TFS, 60, 61-4.

~ Balaban's need for fresh resources with which to reward his own retinue who had supported his

accession.

Our information about the conferment of iqta's is of course more plentiful for the larger kind; and

it is also far greater where Shamsi ghulams are concerned than it is for nobles of free status, about whom

we know very little. Some regions appear regularly as iqta's in Juzjani's Tabaqdt. Iltutmish had at one time

held 'the iqta' of the town (qasaba) of Baran with its dependencies (madafat), for example, and Baran was

frequently granted out as iqta' during his reign and those of his Shamsid successors.61 It is unclear whether

certain major offices automatically carried with them the grant of particular localities. Certainly, a close

link can be detected between the office of wazir and the town of Kol; though this ceased with the overthrow

of Muhadhdhab al-Din in 640/1242.62 In the biography of Aybeg-i Shamsi-yi 'Ajami we read of 'the iqta's

of the armir-i dad, and Juzjani's phrasing suggests at least that in his capacity of dddbeg of the empire that

amir received the iqta' of Palwal and Kama.63 By Kayqubad's reign, there seems to be a clear connection between the office of 'arid and the iqta' of Baran.64

[There are indications that a few major strongholds were 'reserved' (mahrusa) as part of the

sultan's khalisa and were consequently not granted as iqta'. The place most consistently referred to as

mahrusa is Gwaliyor following its recapture in 630/1232-3.65 On that occasion Iltutmish-appointed an

amir-i dad and a castellan (kotwal).66 Subsequently, when its troops were placed under the authority of the

muqta' of Bhayana and Sultankot, he was instructed to make Gwaliyor his headquarters and is said to have

held the intendancy (shihnagi) of that territory (wilayat).61 Certain other important towns are referred to as

mahrusa, but their status appears to have varied. The city of Bihar, for instance, is once termed mahrusa,

but it is also spoken of as an iqta'.68 Tabarhindh is more often described as mahrusa,69 and accordingly we

find Qaraqush Khan appointed as 'intendant (shihna) of the private domain (khalisat) of Tabarhindh'.70 On

one or two other occasions the relationship between the city and its commander is unclear: *Kezlik Khan is described as the 'malik' of Tabarhindh and Arslan Khan is said to have been entrusted with the mahrusa of

Tabarhindh.71 Similarly, although Juzjani applies the term mahrusa to Uchch when

61 TN, I, 443, and II, 8, 21, 25, 27, 29 (tr. 604, 730, 748, 754, 757 [with Bada'un in error], 759).

62 Z. A. Desai, 'Inscriptions of the Mamluk Sultans of Delhi', EIAPS (1966), 8-11; TN, I, 456, 469

(tr. 634, 662). For grantees among the military aristocracy later in the thirteenth century, see ibid., II, 42 (tr.

787); RCEA, XI, 258-9 (no. 4394); TFS, 66, 88, 113.

63 TN, II, 41 (tr. 790-1).

64 TFS, 134, 170.

65 TN, I, 460, and II, 19, 24, 25, 78, 214.

66 Ibid., I, 448 (tr. 620 renders kotwal as 'seneschal').

67 Ibid., II, 10-11 (tr. 732).

68

Ibid., II, 9, 28 (tr. 731; and cf. also 757).

69 Ibid., II, 4, 34, 38, 43, 44 (tr. 723, 767, 784, 792). 70 Ibid., II, 20 (tr. 746).

71 Ibid., I, 446 (tr. 613), for *Kezlik Khan; II, 34 (tr. 767), for Arslan Khan.

~ describing its conquest by Iltutmish in 625/1228 and its conferment on *Kezlik Khan, we read

that in 629/1231-2 'the city and iqta' of Uchch' were granted to Sayf al-Din Aybeg.72 One is tempted to infer

from this wording that the fortified city itself formed no part of an iqtac grant and that whatever authority

the muqta' wielded here rested on a different basis. But this is clearly not the case with Tabarhindh at those

times when it figures unequivocally as an iqta'. Ikhtiyar al-DIn *Altunapa is described as its muqta' under

Radiyya (at a period when it is nevertheless designated as mahrusa);73 'Ala' al-Din Mas'ud Shah granted the

city as iqta' to Ikhtiyar al-Din Yuzbeg (Toghril Khan);74 and later in the reign of the same sultan we read

that 'the fortress of Tabarhindh was assigned to Shir Khan as iqta' and the whole of the dependencies (maddfdt) of the mahrusa of Tabarhindh were bestowed upon him'.75 Such terminology precludes any

possibility that the iqta' grant extended only to the hinterland and did not apply to the stronghold itself.

In some cases the terms in which a grant is couched vary between one recipient and another; and

we even find the same grant described in different language at two different points in the Tabaqdt. Kabir

Khan is said to have been granted 'the city and fortress of Multan, its townships (qasabat) and its districts

near and far (atrdf-u hawali)', and appointed to the governorship (aydlat); elsewhere he is duly termed wall

of Multan.76 Later, however, when held by Qaraqush Khan and a second time by Kablr Khan himself,

Multan is called an iqta'.77 'The territory (wildyat) of Awadh and its dependencies (maddfdt)' were allotted

to Temur Khan by Radiyya, and Qutlugh Khan received the 'government' (aydlat) of Awadh in 653/1255;78

but elsewhere these officers are referred to as muqta's of Awadh, which on other occasions also is expressly said to have constituted an iqta'.79 In rare cases, we gain the impression that the muqta' held the specific

rank of amir-idad in the town entrusted to him, as did Aybeg-i Shamsi-yi 'AjamI in Kasrak and

subsequently in Baran.80 Even in the case of those provinces with more prestigious connotations, like

Lahore, the status of the grant probably varied with that of the grantee. Within a few years, in the hands of

amirs rather than a prince of the blood, Lahore, though still at one point designated as a mamlikat, was

being assigned as iqta'.81

Vague and contradictory terminology prevents us from imposing neat categories on thirteenth-

century arrangements. We clearly cannot expect consistency from our sources, and perhaps in any case the

distinction between gubernatorial status (aydlat) and that of muqta' is of no practical significance. Although

the iqta' was in origin a revenue assignment, the

72 Ibid.;II, 3, 8 (cf. Raverty's tr., 724, 730).

73 Ibid., I, 460; cf. I, 462 (twice).

74 Ibid., II, 30.

75 Ibid., II, 43.

76 Ibid., I, 455-6, and II, 5.

77 Ibid., II, 20, 163. 78 Ibid., II, 17, 69.

79

Ibid., I, 458, 489, and II, 16, 17, 27.

80 Ibid., II, 42 (tr. 791).

81 Ibid., I, 455, 459, 460, 465, and II, 7, 30, 163; also II, 6, for the mamlikat of Lahore.

~

muqta' was not some remote pensionary or military aide at court who had no connection with the

territory in his grant, but an officer who incurred genuine administrative responsibilities. Earlier in the

century Hasan-i Nizami had inserted in his Taj al-Ma'athir the instructions purportedly given to the

unnamed amir who had received the ayalat of the newly .conquered fortress of Banaras. He was to care for

the interests of both the men of the sword and the men of the pen, to protect them from the infidel, to

oversee the labours of the peasants (ra'ayd), to ensure the security of the fortresses, and to discharge the

requirements of charity and good works. The injunctions said to have been issued to Husam al-Din

Oghulbeg at Kol a year or so later are not dissimilar. His duties include not only the waging of jihad, the

guarding of highways and the encouragement of trade, but also honour and preferment to members of the

'religious class' and the administration of justice without distinction between those of good birth and the

common people.82

Some muqta's, if our sources can be trusted, attained these high standards. The Bhayana region

owed its flourishing condition to the efforts of Baha' al-Din Toghril, Mu'izz al-Din's ghulam commander

who had become its first muqta' in 592/1196.83 'Whatever district (ndhiyat) or iqta' or territory (wilayat)

was placed under his control', says Juzjani of Aybeg-i Shamsi-yi 'Ajami, 'has flourished, and the generality

of the subjects ('dmma-yi ra'dya) have been content.'84 The same author assures us that when the two uncles

of 'Ala' al-Din Mas'ud Shah (one of them the future sultan Mahmud Shah) went to their newly conferred

grants, they busied themselves not only with the holy war but also with improving the conditions of the

peasantry.85 Of *Kezlik Khan (d. 629/1231-2) at Uchch, it is said that he strove for the security and repose

of the peasants and performed good works and acts of charity.86 Balaban is praised by Barani for bringing

prosperity to every territory conferred on him as malik or as khan; and Juzjani says that when he first arrived at Hansi he 'gave his attention to cultivation ('imarat), and the people derived contentment from the

monu-ments of his justice and the rays of his generosity'.87 So too another future sultan, Jalal al-Din Khalji,

allegedly caused his assignment of Payal to flourish.88 None of this necessarily testifies, of course, to an

enlightened outlook on the part of the muqta'. He had a vested interest in the material condition of the tract

of which he enjoyed the revenues; and possibly in any case such eulogies, like the instructions cited by

Hasan-i Nizami or the exhortations later ascribed to Sultan Ghiyath al-Din Tughluq,89 tell us at least as

much about what was expected of the muqta' as about what was

82 Taj, fols. 135b-136a, for Banaras; fols. 138a-141a for Kol.

83 TN, I, 421 (tr. 545, 547).

84 Ibid., II, 41 (tr. 789-90 modified).

85 Ibid., I, 470 (tr. 665).

86 Ibid., II, 5 (tr. 724).

87 TFS, 45. TN, II, 52 (tr. 807 modified).

88 FS, 201 (tr. 364).

89 TFS, 430.

~ actually accomplished. Hardy's reminder that the thirteenth-century muqta' was one

'commissioned by the sultan to take charge not of a local territorial unit but of a local situation'90 is surely

salutary. The task of the grantee may have consisted primarily in receiving the tribute from the more

compliant Hindu chiefs at some strongpoint and using it as a base for military operations to extract further

tribute and plunder from their less accommodating peers.91 In a large number of cases, Juzjani can find

little more to say of an amir's activities within a particular iqta' than that he chastised Hindu 'recalcitrants'

(mutamarriddn) and 'rebels' (mufsidan), and destroyed their lairs (mawdsat).92 Few are credited with

founding mosques and implanting Islamic institutions.93

It is possible that in the latter half of the thirteenth century the relative wealth and importance of

an iqta' may have been expressed in terms of the number of horsemen the grantee maintained. Thus Barani

tells us that Balaban's slave Malik *Buqubuq, muqta' of Bada'un, had 4,000 horsemen in his service

(chakir), and that Malik Nusrat-i Sabah, muqta' of Ganuri and Chawpala (now Moradabad) in the reign of

Jalal al-Din Khalji, had 700 horsemen.94 These figures appear modest alongside the capacity of Shir Khan,

who a generation earlier had held Sunnam, Lahore, Deopalpur and 'the iqta's in the path of the Mongol

advance', to raise 'several thousands',95 or the army assembled by Taj al-DIn Sanjar-i Qabaqulaq at Bada'un

in 640/ 1242, which numbered 'eight thousand horse and numerous infantry and paiks' and incurred the

jealousy of unnamed rivals.96

It is to be assumed - though we have little information on the fiscal aspects of the iqta' - that in this

period local revenues outside the khalisa were delivered to the muqta''s appointees rather than to the central government in Delhi. Within his territory the muqta' in turn distributed iqta' grants with a view not only to

recruiting warriors but also to enlisting the administrative capacities of the learned. Among the kindnesses

Juzjani received at the hands of Taj al-Din Sanjar-i Qabaqulaq when he visited Bada'un in 640/1242 was

the bestowal of an iqta'; though 'destiny and fortune' beckoned the chronicler on towards Lakhnawti.97

There seems no reason to doubt that in the Shamsid era the system functioned much as it did later in the

century, under Balaban and Jalal al-Din Khaljl. The muqta' retained so much of the revenue (khardj) from

his grant as he required to pay and fit out his troops, or for other administrative purposes, and

90 Hardy, 'Growth of authority', 203.

91 I. Habib, 'Economic history of the Delhi Sultanate', 295. 92 TN, II, 7-8, 17, 27, 42, 47, 52 (tr. 728, 743, 757, 787, 799, 809). For the mawas, see p. 125

below.

93 TN, II, 26 (tr. 755).

94 TFS, 40, for *Buqubuq, and 204 (reading corrected from BL ms, fol. 110b), for Nusrat-i Sabah.

95 Ibid., 65.

96 TN, II, 26 (tr. 755).

97 Ibid. (tr. 756).

~ remitted the surplus (fawadil) to the capital.98 Barani indicates that by Balaban's era the sultan

nominated an accountant (khwaja) to operate within the province alongside the muqta', reflecting the

government's concern to ascertain the extent of the revenue available.99 Balaban also planted informers

(baridan) in the iqta's to report on the activities of his amirs and their families.100 Barani tells us that

Balaban's son Muhammad personally conveyed the surplus revenue from Sind to his father's court every

year and that he brought three years' revenue following the sultan's return from Bengal.101 We have no

information regarding the proportion of the surplus. In exceptional circumstances, as for particularly

demanding military operations, the muqta' was permitted to keep the surplus also. So it was that in c.

694/1295 the future sultan 'Ala' al-Din was allowed to retain the surplus revenue from his iqta's of Kara and Awadh on the pretext of heading an expedition to Chanderl;102 though in the event he made for Deogir and

won the booty that enabled him to overthrow his uncle the sultan.

What proportion of his time the muqta' was expected or able to spend in his territory is unclear. By

the 1290s we begin to hear of a deputy (na'ib) in some iqta's. Barani's father became both na'ib and khwaja

of Baran at 'Ala' al-Din's accession.103 His responsibilities, presumably, included the super-vision of

revenue collection and the fitting out of troops. A muqta' with major administrative commitments in Delhi

was evidently not expected to see to these matters in person. Juzjani, describing how Ulugh Khan Balaban

in 653/1255 had to go to Hansi to oversee the mustering of contingents from the Siwalik region (Hansi,

Sarsati, Jind and Barwala), which had been subject to delay, appears to suggest that this was unusual.104

Those who fell foul of the government could suffer banishment to their iqta's, a fate met with

relatively frequently in the pages of the Tabaqdt-i Nasiri. To draw examples from the power struggle of the

1250s, Ulugh Khan Balaban, when he forfeited the office of na'ib in 651/1253, was ordered to retire to

Hansi; in 653/1255 his enemies Rayhan and Qutlugh Khan were dismissed respectively to Bada'un and to

Awadh; and in 655/ 1257 any of their confederates among the religious aristocracy of Delhi who held an

iqta' in the vicinity (hawati) of the capital were ordered to take up residence there.105 Almost four decades

later, certain of the amirs accused of conspiring against Sultan Jalal al-Din Khalji were punished by being

sent to their iqta's for one year, a penalty which Barani evidently intended his

98 TFS, 164, 220-1. I. Habib, 'Agrarian economy', in Raychaudhuri and Habib, 69, more

cautiously, suggests that the despatch of the surplus from the iqta's was being demanded 'well before the fall of Balban's dynasty'.

99 TFS, 36-7 (in the iqta' of Amroha); and see I. Habib, 'Agrarian economy', 69-70.

100 TFS, 40, 44-5.

101 Ibid., 69, 108-9.

102 Ibid., 220-1.

103 Ibid., 248. 104 TN, II, 70-1 (tr. 837); see also I, 490 (tr. 703).

105 Ibid., I, 486, 489, 492 (tr. 693, 700, 701, 708); also II, 64, 69 (tr. 826, 834).

~ readers to view as excessively lenient.106 Bada'un often figures as the destination of those who

lapsed from favour at court, as it was for Badr al-Din Sonqur in 638/1241, for Kushlu Khan in 649/1251

and for Rayhan three years later (in each case, as its newly appointed muqta'), and for the deposed qadi

Imad al-Din Shafurqani in 646/1248.107 The fate of Qadi Jalal Kasani, who was implicated in the Sidi

Muwallih affair, was to be sent off to Bada'un as qadi.108 Under Sultan Balaban the penalties were harsher.

His slave Malik *Buqubuq, sar-i jdnddr and muqta' of Bada'un, was executed for slaying a chamberlain

(farrash), and the qorabeg Haybat Khan, who held the iqta' of Awadh, narrowly escaped the same sentence for likewise killing a man.109 But perhaps these were the exceptions.

Evidence from the period of Ulugh Khan's ascendancy as na'ib, from 653/1255 onwards, suggests

that certain grants and offices were becoming hereditary, especially those held by the religious aristocracy,

who tended to enjoy incomes exempt from any kind of service and known as in'amdt. Juzjani received

grants in this category from Ulugh Khan Balaban, namely a village in the Hansi region, of which he took

possession in 647/1249-50, and another village (in an unspecified location), together with a pension in

cash, on the completion of his Tabaqdt.110 When the Shaykh al-Islam Jamal al-Din Bistami and the qadi

Kabir al-Din died in 657/1259, their offices (manasib) were conferred on their sons; and in like fashion the

in'dmat of the imam Hamid al-Din of Marigala, who died a few months later, passed to his children. It

comes as a greater surprise to find this trend affecting military and administrative office. Kishli Khan was succeeded as amir-hajib in this same year by his son 'Ala' al-DIn Muhammad, and in Balaban's reign the

'arid was followed by his son.111 In the early Khalji era Taj al-Din 'Iraqi transmitted his office of omir-i

ddd-i lashgar to his son Kabir al-DIn.112 Apropos of iqta's, however, the hereditary principle carried less

weight. We know of no muqta' whose grant passed on his death to a relative; but Juzjani reveals that Arslan

Khan was deemed to merit the iqta' of Bhayana on the grounds that he had married the daughter of Baha'

al-Din Toghril, its first muqta'.113 In the reign of Nasir al-Disn Mahmud Shah, Qilich Khan Jalal al-Din

Mas'ud, 'Ala' al-DIn Jam's son (see appendix II), was twice granted Lakhnawti, which his father had briefly

106

TFS, 192.

107 Sonqur: TN, I, 465, and II, 25 (tr. 654, 753). Others: ibid., I, 482, 485, 489, and II, 68 (tr. 685,

690, 700, 833).

108 TFS, 211.

109 Ibid., 40-1, 84.

110 TN, I, 481, 484, and II, 61, 220 (tr. 681, 687, 821-2, 1294-5).

111 Ibid., I, 495 (tr. 713); for' Ala' al-Din as amir-hajib under Balaban, see also TFS, 24, 35, 36,

113. TS, IOL Persian ms. 412, fol. 36b.

112 TFS, 361.

113 TN, II, 34 (tr. 767). It is also possible (though Juzjani makes no explicit link here) that Temur

Khan's claim to Lakhnawti, when he wrested it from Toghan Khan in 642/1244, derived from his marriage

to a daughter of its former muqta', Sayf al-Din Aybeg-i Yaghantut: ibid., II, 18 (tr. 744).

~ eld in iqta' under Iltutmish.114 There is perhaps meagre evidence that the ereditary principle was

not totally irrelevant in the allocation of major qta's in the thirteenth century, several decades before the

Tughluqid Sultan iruz Shah made it the chief criterion.

4 Ibid., II, 78 (tr. 848-9); RCEA, XI, 211 (no. 4320). For Jani, see TN, I, 448, and II, 9 (tr. 618,731-

2).

~CHAPTER 6

The Mongol threat

The Mongol world-empire

When Chinggis Khan died in 1227, without having returned to western Asia, his empire extended

from the steppes of present-day Mongolia to north-eastern Persia and the Hindu Kush. It was not until the

election of his son and successor, Ogodei, as great khan (qaghanlqa'an), at an assembly (quriltai) in

Mongolia in 1229, that the Mongols again paid any attention to the Indian borderlands, and only in 639/1241 that they first entered the territory of the Delhi Sultan, thus inaugurating a long period of

hostilities with the Sultanate which lasted beyond the sack of Delhi by the Central Asian conqueror Temur

in 801/1398. These conflicts are less fully covered than Chinggis Khan's own invasion of India. The

principal sources for Mongol activities in eastern Persia and Central Asia - Juwaynl's Ta'rikh-i Jahan-

Gusha (658/1260), Rashid al-Din's Jami' al-Tawarikh (c. 703/1303-4) and Wassaf's Tajziyat al-Amsar -

devote comparatively little attention to Mongol relations with the subcontinent. As we have seen, Wassaf's

work includes also a brief history of the Delhi Sultanate down to the early years of 'Ala' al-Din Khalji,

which was in turn utilized by Rashid al-Din and by Qashani (c. 718/1318). Details of a few Mongol

campaigns in India can also be gleaned from the history of Herat (Ta'rikh-Nama-yi Harat) by Sayfi (c.

722/1322). But these Iranian historical traditions only partially fill in the gaps left by Indo-Muslim

chroniclers like Juzjani and Barani, who report Mongol invasions but conversely display virtually no interest in conditions within the Mongols' own territories.

By Ogodei's death the Mongols had adopted an ideology of world conquest, according to which

the whole earth was already granted to them by the eternal sky-god (Tenggeri). Other rulers had a clear

duty to recognize their place within this world-empire, to submit in person to the qaghan, to put their troops

at his disposal, to accept a Mongol resident (shihna) and to dismantle their fortifications. Any restraint the

Mongols manifested in making good their title to the whole world sprang only from tactical considerations.

It was sometimes necessary to make a truce with one ruler

103

~ in order to concentrate on another enemy elsewhere. But in the Mongols' vocabulary, 'peace' and

'submission' were the same word (Tu. ill): the qaghan had no allies, only subjects.1

As we saw (p. 34), the Mongols' principal attacks in 620-1/1223-4 were directed against Nandana

and the Lahore region, then under Khwarazmian domination, and Multan, which belonged to Qubacha.

Professor Habi-bullah suggested that Chinggis Khan refrained from further operations in India out of

regard for the Delhi Sultan's neutrality, as demonstrated in his failure to assist Jalal al-Din. The Mongol

conqueror is further credited with 'moderation' and a 'scrupulous observance of international practice'.2 If

so, Chinggis Khan's policy towards India in 1223 affords a unique instance of this spirit. There is, in any

case, no reason why he should have regarded India as an immediate objective, on a par with the empire of

the Khwarazm-shah. At this time, Khwarazm, Transoxiana and the Ghazna region had yet to be pacified, and while he was based south of the Hindu Kush divisions of his army were engaged in vital campaigns to

suppress revolts in Khurasan. Even the troops that Ogodei later sent to this region were designated in the

so-called 'Secret history of the Mongols' merely as a reserve force (Mo. gejige) for the main army operating

in Persia under Chormaghun.3

It is possible that Iltutmish made some gesture of submission of which Juzjani does not tell us.

Ninety years later, in 710/1310-11, in an embassy to Sultan 'Ala' al-DIn Khalji, the Ilkhan Oljeitii reminded

him how his predecessors had, 'both in the time of Chinggis Khan and in the time of ... Ogodei Qa'an,

breathed the breath of conciliation and obedience and through the words of envoys had laid the

countenance of loyalty on the face of the earth'.4 This may, of course, amount to no more than diplomatic

swagger. The arrival of 'the ruler of In-tu (Hind)' at Ogodei's court, reported under the year 1229 in the Chinese dynastic history of the Mongol period, the Yuan Shih,5 perhaps refers to some Hindu prince or to

Hasan Qarluq, the ruler of Binban. Yet we know that Iltutmish did receive embassies from the Mongols,

since JuzjanI assures us that he never

1 Eric Voegelin, 'The Mongol orders of submission to European powers, 1245-1255', Byzan-tion

15 (1940-1), 378-413. Klaus Sagaster, 'Herrschaftsideologie und Friedensgedanke bei den fruhen

Mongolen', CAJ 17 (1973), 223-42. Igor de Rachewiltz, 'Some remarks on the ideological foundations of

Chingis Khan's empire', PFEH 1 (1973), 21-36. David Morgan, 'The Mongols and the eastern

Mediterranean', in B. Arbel et al. (eds.), Latins and Greeks in the eastern Mediterranean after 1204

(London, 1989 = Mediterranean Historical Review 4, no. 1), 200, suggests that this ideology postdated

rather than preceded the burst of expansion under Chinggis Khan.

2 Habibullah, Foundation, 206.

3 Mangghol un niuca tobca'an, para. 270, ed. (in transcription) L. Ligeti, Histoire secrete des

Mongols (Budapest, 1971), 243; tr. F. W. Cleaves (Cambridge, Mass., 1984), 210; tr. Igor de Rachewiltz, in

PFEH 31 (March 1985), 26 and n. at 58.

4 Wassaf, 528.

5 Waltraut Abramowski, 'Die chinesischen Annalen von Ogodei und Guyuk', ZS 10 (1976), 125.

~ killed their envoys but simply sent them off under guard in some fashion (ba-tariqi).6

The first Mongol encroachments

The reign of Ogodei (626-39/1229-41) witnessed a steady build-up of pressure beyond the frontier

of the Sultanate. After the first quriltai, troops under Dayir Noyan advanced from Herat into Sistan and

overthrew its ruler, Yinaltegin (632/1235).7 It was probably the Mongol forces in Tukharistan, Qunduz and

Taliqan, as also, it seems, those in Ghazna, together totalling two tumens (20,000), under the command of

Monggedu (Mengutei),8 which soon afterwards moved into Kabul, Ghazna and Zabulistan and obliged

Hasan Qarluq to accept a Mongol resident (shihna).9 At a second quriltai in 632/1235, further Mongol

troops under *Oqotur were ordered to advance on India, and Kashmir was ravaged in the course of a campaign lasting six months.10 In 636/1238-9 Qarluq, who had become tributary to the Mongols, was

suddenly attacked by the generals (noyans) Anban and Neguder and expelled from his territories of

Ghazna, Kurraman and Binban. He fell back on the Sultanate, launching an attack on Uchch which was

repulsed by its muqta', Sayf al-Din Aybeg.11 The Mongols' campaign against Hasan Qarluq brought them to

the frontiers of the Delhi Sultanate, and they now occupied the territories which had served as the

springboard for the Ghurid invasions of India two generations earlier. In 639/1241 an army under the joint

leadership of Dayir and Monggedii invested Lahore. The muqta', Qaraqush Khan, fled, and although Dayir

was killed in the fighting the city fell on 16 Jumada 11/22 December 1241:12 the reaction from Delhi,

where Bahram Shah was highly

6 TN,II,214(tr. 1284).

7 Bosworth, History of the Saffarids, 409-10. For Dayir, see Boyle, 'Mongol commanders', 240;

Aubin, 'L'ethnogenese', 70-1.

8 TN, II, 153, 169 (tr. 1109, 1152). JT, I, part 1, ed. A. A. Romaskevich et al. (Moscow, 1965),

188, and tr. A. A. Khetagurov, Sbornik letopisei, I, part 1 (Moscow and Leningrad, 1952), 109 (where the

subject of the sentence is wrongly taken to be Mongke), similarly describes the camping-grounds of this

army as 'Qunduz-i Baghlan and the confines (hudud) of Badakhshan'. For the form of Monggedu's name,

see Boyle, 'Mongol commanders', 242 and n.67. On the tiimen, a unit of (sometimes notionally) 10,000, see

D. O. Morgan, The Mongols (Oxford, 1986), 89.

9 TN, II, 159 (tr. 1119). 10 JT, II, 42 = II, part 1, ed. A. A. Alizade (Moscow, 1980), 120 (tr. Boyle, 55/tr. Verkhovskii, 36).

JT, ed. Jahn, Indiengeschichte, Ar. text Taf. 61 (German tr. 56); there is a lacuna in the TSM ms. here. The

BL Persian ms. Add. 7628, fol. 391a, says that the Mongols stayed six months and that the raja of Kashmir

returned after seven years. This error misled Jahn, 'A note on Kashmir and the Mongols', CAJ 2 (1956),

177; his date for the invasion (ibid., 179) is also wrong.

11 TN, II, 8-9, 162 (tr. 730, 1128-9).

12 Ibid., II, 163-5 (tr. 1133-6); II, 166 (tr. 1142), for the date, which is given as Jumada I at I, 465-6

(tr. 655); for the joint command, see II, 6 (the meaning of dar muwafaqat is obscured in Raverty's tr., 727).

Raverty (1135 n.5) rejects Juzjani's testimony on the grounds that

~ unpopular with the military, was ineffectual. In 643/1245-6 a campaign by Monggedii dislodged

from Multan Hasan Qarluq, who had recently seized the city, and forced him to flee by boat down the Indus

towards Siwistan and Daybul; the Mongols then invested Uchch for a time, before an expedition by Sultan

'Ala' al-Din Mas'ud Shah obliged them to retreat. Uchch and Multan were pacified by a detachment under

Aybeg-i Khita'i the sar-ijandar and muqta' of Baran.13

JuzjanI gives prominence in his account of this campaign to the role of the future sultan Balaban

in securing Monggedii's withdrawal, and proudly describes how the Delhi forces went over to the offensive

in the winter of 644/1246-7, early in the reign of Nasir al-Din Mahmud Shah, when Balaban advanced as

far as the Indus, greatly intimidating the Mongol frontier patrols. But the chronicler's statement that the Mongols were thereby deflected from invading in this year betrays the fact that their inroads had become an

annual event.14 In any case, they soon sought to take advantage of the fighting around Multan between

Kushlu Khan and his rival Shir Khan (see pp. 71-2), for JuzjanI records the despatch of a great number of

Mongol prisoners to court in 648/1250-1 by *Kiirbuz, Shir Khan's deputy at Multan.15

Tensions within the Mongol empire

Mongol operations on the Sultanate's western frontier since 639/1241 had proved less than

impressive. Why, we cannot be certain. Dayir's death would account only for the immediate withdrawal

from Lahore. The notions that the climate made India an unattractive goal and that the Panjab was ecologically unsuited to the Mongols' own brand of pastoral nomadism must be discarded. It is true that the

heat had compelled Dorbei to withdraw from Multan in 621/1224;16 but this would hardly explain the

failure of the Mongols to establish themselves in the Panjab. Not only were they accustomed to climatic

extremes in their original habitat, but in the fourteenth century we find them wintering in India on a regular

basis, which suggests that they found adequate pasturage for their livestock. The aims of the Mongol

invasions of India will be discussed more fully in chapter 11.

More important in the longer term were the tensions within the imperial dynasty. Firstly, the lines

between the qaghan's sphere of authority and those of his more important kinsfolk were increasingly

blurred. On the one hand, Chinggis Khan had allotted to each of his relatives a specific

Dayir is mentioned as still alive in Mongke's reign (1251-9). But this is a misunderstanding based on the vagueness of Rashid al-Din: see JT, III, 21-2 (tr. Arends, 22).

13 TN, I, 471, and II, 28, 54-6, 170-1 (tr. 667-8, 758, 809-13, 1153-6).

14 Ibid., I, 479-80, and II, 56-7 (tr. 677-9, 814-16).

15 Ibid., I, 484 (tr. 688); for the appointment of *Kurbuz to Multan, see II, 38, 44 (tr. 782, 792).

16 TJG, I, 112 (tr. Boyle, 142).

~ pasturage together with a certain number of nomadic subjects - the complex termed in Mongolian ulus. Juwayni describes the largest of such units, those granted to the conqueror's four sons,

Jochi, Ogodei, Chaghadai and Tolui, as radiating out from the homeland in Mongolia in a westerly

direction according to seniority. As the eldest, Jochi was entrusted with the westernmost territory 'as far as

the hooves of Mongol horses had trodden';17 and when he died, shortly before Chinggis Khan, his son Batu

became the real founder of the Mongol power in the Pontic and Caspian steppes, known to historians as the

Golden Horde.

On the other hand, the Mongol conquests were regarded as the joint possession of the entire

imperial family. The sedentary regions appear from Ogodei's reign onwards to have been run by 'satellite

administrations' comprising representatives of both the qaghan and neighbouring princes.18 The principle of

joint rule found expression also in the tama system, which was elucidated by the late Jean Aubin. The

forces sent to each newly conquered territory included contingents furnished by each branch of the imperial family, so that the interests of one prince could not be furthered without the consent of relatives whose

forces were operating alongside his own. Thus in the Indian borderlands in the 1230s the families of

Chinggis Khan's four sons were each represented by a commander.19

We should note, secondly, the absence of recognizable rules for the succession to the qaghanate. A

prince designated by the previous monarch was often ignored in favour of a more competent or senior

member of the dynasty, a system for which the late Professor Joseph Fletcher borrowed from the Celtic

world the label 'tanistry'.20 Thus during the five-year interregnum that separated the death of the qaghan

Ogodei in Mongolia in December 1241 from the election of his son Giiyiig (644-6/1246-8), rumours

reached Delhi of bitter conflicts among the princes, including the sons of Chinggis Khan's younger brother

Temiige.21 Guyug's early death, which averted a fratricidal war with his cousin Batu in Central Asia, was followed by another interval of three years, after which Batu and his party elbowed aside Giiyiig's offspring

and secured the imperial dignity for Tolui's son Mongke (1251-9). Opposition from the two middle

branches of the dynasty was ruthlessly crushed; in a major redistribution of resources, the majority of the

Ogodeyid and Chaghadayid princes were deprived of their pasturelands and were put to death or exiled.22

To rally support behind the new regime, campaigns were instituted on

17 Ibid., 1,31 (tr. Boyle, 42).

18 Paul D. Buell, 'Sino-Khitan administration in Mongol Bukhara', JAH 13 (1979), 141-7. 19 Sayfi, 174. For the tama system, see Aubin, 'L'ethnogenese', 74-5.

20 J. Fletcher, 'The Mongols: ecological and social perspectives', HJAS 46 (1986), 17, 24-8.

21 TN, II, 166-7 (tr. 1143-4). See P. Jackson, 'The dissolution of the Mongol empire', CAJ 22

(1978), 197-9.

22 Ibid., 200-8.

~ fronts as far distant as China and Western Asia.23 In 1252-3 Sali Noyan was sent to the Indian

borderlands at the head of fresh troops, and was given authority over all the forces commanded in the past by Dayir and by *Oqotur. Sali was himself subordinated to Mongke's brother Hulegu,24 who in 653/1255

was to march westwards at the head of a great army and would crush both the Isma'ili Assassins in northern

Persia (654/1256) and the 'Abbasid Caliphate at Baghdad (656/1258). In accordance with the tama system,

the qaghan's forces were accompanied by contingents representing other princes of the imperial dynasty,

among which the troops supplied by the Jochids predominated.25

The distant location of the Jochid ulus gave its princes a good deal of practical independence from

the qaghan. Batu's power extended well beyond the confines of what would later be Horde territory: Juzjani

testifies to his authority throughout those parts of Persia occupied by the Mongols, and Sayfi furnishes

more specific evidence, showing how Batu intervened in the affairs of Herat in the 1240s.26 According to

reports which reached Egypt ten years or so later, the Golden Horde was entitled to anything from a third to two-fifths of the spoils from Persia.27 But although Hiilegii's presence in Western Asia might have been

seen as a challenge to the Jochids' position there, friction arose only after the death of Mongke in 1259 and

the outbreak of a struggle in the following spring between his brothers Qubilai and Arigh Boke in the Far

East.

Even as he ceased writing, JuzjanI had heard the first rumours of tension between Hulegii and his

cousin Berke, Batu's brother and now ruler of the Golden Horde. The reason given - the outrage felt by

Berke, who had been reared as a Muslim, at the fate of the 'Abbasid Caliph - is also found in sources from

the Mamluk empire.28 In fact, however, conflict between Berke and Hiilegu was deferred for some three

years following the sack of Baghdad, and authors writing further west, who were better placed to observe

events, attribute the clash to two quite different causes. One is that Hulegii deprived the Golden Horde of

its customary share of the spoils from Persia. The other reason furnished in the sources for the enmity between Hulegii and Berke is that they supported rival candidates in the

23 T. T. Allsen, Mongol imperialism: the policies of the Grand Qan Mongke in China, Russia, and

the Islamic lands, 1251-1259 (Berkeley and Los Angeles, 1987), 1-5.

24 JT, I, part 1, 188 (tr. Khetagurov, 110); JT, III, 21-2 (tr. Arends, 21). Aubin, 'L'ethnoge-nese',

73, 79. Sali's despatch to India is dated in July 1253 by the Yuan Shih: Waltraut Abramowski, 'Die

chinesischen Annalen des Mongke', ZS 13 (1979), 22.

25 Jackson,'Dissolution', 220-1.

26 TN, II, 176 (to be corrected from BL ms., fol. 155b; cf. Raverty's tr., 1172). Sayfi, 122-8, 136-9.

For Batu, see Jackson, 'Dissolution', 212-13, 218.

27

D. Ayalon, 'The Great Yasa of Chingiz Khan: a re-examination', part B, Studia Islamica 34

(1971), 174, repr. in his Outsiders in the lands of Islam (London, 1988). Jackson, 'Dissolu-tion', 220-1.

28 TN, II, 198 (tr. 1257). al-Yuninl, II, 365. al-Safadi, Wafi, X, 118.

~ succession dispute: Berke acknowledged Arigh Boke, whereas Hulegu favoured Qubilai.29

Hulegu's precise status at the time of Mongke's death is problematic. When describing the

terms of his commission, the Ilkhanid chronicler Rashid al-Din uses highly guarded language.

According to his version of events, Mongke privately intended his brother to remain in Persia and

transmit it to his descendants (the so-called 'Ilkhans'), but he made a show of ordering him to

return to Mongolia once the conquest was completed.30 This reads suspiciously like an attempt to

justify the position of the Ilkhans retrospectively. It must be emphasized that we have no other

evidence for any such purpose on Mongke's part, and sources composed within the Mamluk

empire assert, on the contrary, that at some point after the fall of Baghdad Hiilegii rebelled and

established himself as the ruler of the province.31 Even the actual title (il el, 'subordinate', khan)

taken by Hulegii and his line is not attested prior to 658/1260.32 He appears to have profited from

the outbreak of conflict in the Far East to convert his position from that of commander-in-chief in

Persia to that of ruler of an ulus on a par with his kinsmen, receiving from Qubilai the legitimation he so needed. The Jochid princes and generals in his army were arrested and executed or

imprisoned, and most of their troops slaughtered; and he was then free to encroach on the

territories south of the Caucasus that the Jochids regarded as their own. As a consequence, war

broke out in 1261.33

The events following Mongke's death marked the dissolution of the Mongol empire. Even after

Arigh Boke's submission in 1264, Qubilai, who reigned from the new capital of Khanbaligh (Ta-tu) in

northern China, could count on the allegiance only of Hiilegii and his descendants, the Ilkhans of Persia, so

that it is possible to speak of a 'Toluid axis' comprising these two geographically remote powers. From c.

1270, moreover, he was confronted with a coalition of enemies in Transoxiana and Turkestan. Here the

Chaghadayid prince Alughu, who had defected to Qubilai from Arigh 29 John W. Dardess, 'From Mongol empire to Yuan dynasty: changing forms of imperial rule in

Mongolia and Central Asia', Monumenta Serica 30 (1972-3), 128-31. P. Jackson, 'The accession of Qubilai

Qa'an: a re-examination', Journal of the Anglo-Mongolian Society 2 (1975), 1-10. Idem, 'Dissolution', 226-

7'. Morris Rossabi, Khubilai Khan: his life and times (Berkeley and Los Angeles, 1988), 46-62.

30 JT, III, 24 (tr. Arends, 22).

31 MA, ed. and tr. Klaus Lech, Das mongolische Weltreich (Wiesbaden, 1968), Ar. text 2 (German

tr. 91).

32 Reuven Amitai-Preiss, 'Evidence for the early use of the title il-khan among the Mongols',

JRAS, 3rd series, 1 (1991), 353-61. Thomas T. Allsen, 'Changing forms of legitimation in Mongol Iran', in

G. Seaman and D. Marks (eds.), Rulers from the steppe: state formation on the Eurasian periphery (Los

Angeles, 1991), 227.

33 For a more detailed investigation of the origins of this conflict, see Jackson, 'Dissolution', 226-

35; also idem, 'From ulus to khanate: the making of the Mongol states, c. 1220-1290', in R. Amitai-Preiss

and D. O. Morgan (eds.), The Mongol empire and its legacy (forth-coming).

~ Boke, had nevertheless profited from the civil war to re-establish Chagha-dai's ulus on quite new

foundations, appropriating for himself the revenues of the neighbouring sedentary regions which should

have gone to the qaghan; Baraq, whom Qubilai despatched west to rule Chaghadai's ulus after Alughu's death, soon defied him. But the most dangerous enemy confronting Qubilai in this region was Qaidu, a

grandson of Ogodei, who was recognized as their khan by a number of Chaghadayid and Ogodeyid princes

and noyans in 670/1271 when Baraq died.34 Qaidu's empire was an extensive one. He took over the fiscal

administration of the sedentary regions of Central Asia, whose officials were now his appointees, and he is

found nominating the rulers of Chaghadai's ulus, who seem to have acted as his subordinates. He and his

allies remained hostile to the regime at Khanbaligh until his death in 1303; and the Mongol world did not

acknowledge a single qaghan again until 1304 (below, p. 220).35

The disintegration of their empire into a number of rival khanates seriously impaired the Mongols'

capacity to prosecute expansionist cam-paigns on any front, whether in China, in eastern Europe, in Syria or in India. The Ilkhans were required to keep vigilant watch for an invasion of Transcaucasia by the forces

of the Golden Horde. Periodic attacks by the Chaghadayid Mongols, and particularly that of 668/1270,

effectively turned Khurasan at times into a no-man's land.36 The Ilkhans retaliated in 671/1272-3 by

sacking Bukhara.37 The abandonment of the old claim to world-rulership is most starkly demonstrated in

the new-found readiness of Mongol princes to ally with outside powers against their own kinsfolk. The

Ilkhans were confronted from 662/1263-4 by an understanding between their northern neighbours, the

Golden Horde, and the Mamluk regime in Egypt and Syria. Their own efforts to counteract this by

negotiating for joint action with the Mamluks' enemies in Catholic Europe were unavailing.38

34 Jamal al-Qarshi, Mulhaqat al-Surah, ed. in V. V. Bartol'd, Turkestan v epokhu mongol'skogo

nashestviia (St. Petersburg, 1898-1900, 2 vols.), I (texts), 138. Wassaf, 76. JT, II, 192 (tr. Boyle, 153/tr.

Verkhovskii, 100), and III, 138 (tr. Arends, 87). 35 The fullest survey of Qaidu and his empire is to be found in Michal Biran, Qaidu and the rise of

the independent Mongol state in Central Asia (Richmond, Surrey, 1997). Briefer accounts in W. Barthold,

Four studies on the history of Central Asia (Leiden, 1956-62, 4 parts in 3 vols.), I, 124-9; Pelliot, Notes on

Marco Polo (Paris, 1959-73, 3 vols.), I, 124-9; J. A. Boyle, 'Kaydu', Enc.Isl2; more generally, Dardess,

130-1.

36 JT, III, 148 (tr. Arends, 92).

37 Berthold Spuler, Die Mongolen in Iran, 4th edn. (Leiden, 1985), 63.

38 S. Zakhirov, Diplomaticheskie otnosheniia Zolotoi Ordy s Egiptom (Moscow, 1966). Reuven

Amitai-Preiss, Mongols and Mamluks: the Mamluk-Ilkhdnid war, 1260-1281 (Cambridge, 1995), 78-86,

94-105. J. A. Boyle, 'The Ilkhans of Persia and the princes of Europe', CAJ 20 (1976), 25-40. Morgan,

Mongols, 183-6. For a more detailed survey of the halting of the Mongol advance in 1260-2, see Jackson,

'Dissolution', 236-43.

~ The Sultanate's 'Mongol crisis'

The build-up of Mongol power in Persia had been all the more menacing in view of the fact that

the Mongols were being drawn into the internal affairs of the Delhi Sultanate following the flight into

Mongol territory in 646/1248 of Mahmud Shah's brother Jalal al-Din Mas'ud. The sources composed in

Persia state that Mongke ordered Sali Noyan to assist him to recover 'his ancestral realm'. Successive attacks by Sali on Multan and Lahore, which are described by Sayfi and in which the governor of Multan

bought off the invaders, seem to have formed part of this effort. Regarding the fate of Lahore we are told

nothing, since Sayfi breaks off at this juncture to describe the fortunes of Shams al-Din Muhammad Kart,

the client malik of Herat, who had accompanied Sali's forces but now withdrew and returned home. The

expedition of Mahmud Shah's forces towards Sind by way of Lahore in 650/1252, in which Ulugh Khan

Balaban fell from favour (see p. 72), must have been a response to this inroad.39 As a result, the Mongols

were unable to penetrate further than Jajner and fell back, but the prince was installed as client ruler of

Lahore, Kujah and Sodra, which were subject (il) to them.40

Jalal al-Din Mas'ud's authority in these regions did not last long after his participation in the

campaign which restored Ulugh Khan Balaban to power in 652/1254. At some point soon afterwards, he was joined by Balaban's cousin Shir Khan, who had also taken refuge in Mongol territory and returned in

the wake of the na'ib's reinstatement. The two men fell out, and Jalal al-DIn, who retired and left his

dependants and troops in the hands of his rival, probably died within the next few years. When Shir Khan

engaged in conflict with Arslan Khan, the muqta' of Tabarhindh, the Delhi government intervened and

granted him, in return for his allegiance, not only Tabarhindh but 'the whole of the territory and iqta's which

he had previously held': this formula must have been designed to embrace Lahore, Uchch and Multan.41

39 TN, I, 486 (to be corrected from BL ms., fol. 192a; cf. Raverty's tr., 692). Sayfi, 157-9 (sub anno

644). Habibullah (Foundation, 215) was thereby led to connect the invasion with Balaban's campaign in the

Salt Range. But Sayfi's chronology is unreliable for this period (see Aubin, 'L'ethnogenese', 72-3), and the year is too early for either Sali or Shams al-DIn Kart to be in the Indian borderlands. That Sayfi mentions

the 'ld-i Qurban (10 Dhu'l-Hijja) as falling during the siege of Multan (158) is an argument in favour of

locating this expedition in 650, where it harmonizes with the time of year specified by Juzjani. The

governor of Lahore, whom Sayfi names as KRT Khan (Kirit Khan?), cannot be identified. We should,

perhaps, have expected Toghril Khan (Yiizbeg): see above, p. 92.

40 Wassaf, 310; JT, ed. Jahn, Indiengeschichte, Ar. text Taf. 57, Pers. text Taf. 22 (tr. 48 reads

'Haibar' for the HHNYR of the mss.); Qashani, 185. For this episode, see Karl Jahn, 'Zum Problem der

mongolischen Eroberungen in Indien (13.-14. Jahrhundert)', in Akten des XXIV. internationalen

Orientalisten-Kongresses Munchen ... 1957 (Wiesbaden, 1959), 618.

41 TN, II, 44 (tr. 793). This settlement must have occurred prior to 654/1256, when we find Arslan

Khan in Awadh. That Jalal al-DIn Mas'ud was dead by 658/1260 is indicated by the

~ Mongol interests in the Lahore region had already suffered a setback, it is fair to assume, as a

result of Jalal al-Din Mas'ud's reconciliation with Mahmud Shah.42 Now he had been supplanted by Shir

Khan, who was subject to the sultan and, encouraged by Delhi, harboured designs on Sind. This region too

had recently become a Mongol protectorate. Following his triumphant return to Uchch and Multan in

651/1254 Kiishlii Khan had used the good offices of Shams al-Din Muhammad Kart of Herat to offer his

submission and had accepted a Mongol shihna. When he and his confederates failed to take Delhi in

655/1257 (see pp. 74-5), Kiishlii Khan turned to his Mongol overlords. His appeal to Hulegu for assistance

-made, according to Juzjani, in person - elicited an immediate response. In the winter of 655/1257-8 Sali

Noyan entered Sind in strength and dismantled the fortifications of Multan; his forces may also have invested the island fortress of Bhakkar on the Indus.43 Although the sultan's army moved out of Delhi, its

stance appears to have been purely defensive. The government was evidently concerned not to provoke the

Mongols. In Safar 657/February 1259 Shir Khan was transferred to an extensive assignment centred on

Bhayana and hitherto held by Nusrat Khan, who now replaced him at Tabarhindh, and Juzjani expressly

ascribes the exchange to the need to avert conflict on the frontier, presumably with Kushlu Khan.44 The

impression of a propitiatory attitude is heightened by the continued failure of the Delhi forces to take action

against the Mongols during these months, while the enemy assailed the sultan's territory.

When Juzjani wrote, the Mongols had overwhelmed 'the whole of the land of Turan and the east'.

Everywhere 'from the borders of China, Turkistan, Ma wara' al-Nahr, Tukharistan, Zawul[istan], Ghur,

Kabul, Ghaznayn, Iraq, Tabaristan, Arran, Adharbaijan, the Jazira, Anbar, Sistan, Makran, Kirman, Fars,

Khuzistan, Diyarbakr, and Mawsil, as far as the limits of Rum and Syria', Muslim rulers had been swept away.45 During the previous two decades, the territories that owed allegiance to the Delhi Sultan had shrunk

dramatically under the impact of the Mongol advance. Iltutmish's operations against the Khwarazmshah's

lieutenants, which we noticed earlier (p. 36), had brought under his control a number of important

formula 'alayhi'l-rahmatu in Juzjani's list of Iltutmish's offspring in BL ms., fol. 179b, and IOL

ms., fol. 242b.

42 TN, I, 489 (tr. 700), says that Lahore was recognized as his iqta'.

43 Ibid., I, 494, and (with lashgarha in error for kungurha) II, 76 (tr. 711, 844); for Kiishlii's

submission, see II, 38-40 (tr. 784, 786). Bhakkar: Sayfi, 250-7 (sub anno 657). Kiishlii Khan's dealings with the Mongols are perhaps linked with the embassy from 'a sultan of India' which accompanied a

Western European missionary on the early stages of his journey from Mongke's court in July-August 1254

(although this could equally refer to Shir Khan): William of Rubruck, 'Itinerarium', xxxvi, 3, in A. Van den

Wyngaert (ed.), Sinica Franciscana, I. Itinera et relationes Fratrum Minorum saeculi XIII et XIV

(Quaracchi-Firenze, 1929), 306; tr. P. Jackson and D. O. Morgan, The mission of Friar William of Rubruck,

HS, 2nd series, 173 (Cambridge, 1990), 247 and n.2.

44 TN, II, 44 (tr. 794). 45 Ibid., II, 90-1 (tr. 870-1, 879-85).

~ strongholds between the Jhelam and the Ravi. Juzjani lists among his conquests Kujah (the modern Gujrat), Nandana, Sodra and Siyalkot; Kujah and Nandana are described as border regions

(sarhadd).46 But in Radiyya's reign the Mongols already held the tracts beyond the Chenab, which is

doubtless why the rebel Kabir Khan, when pursued north by Radiyya's forces in 637/1239, was unable to

retreat further than 'the confines of Sodra'.47 By the time the renegade prince Jalal al-Din Mas'ud returned

with Mongol aid in c. 650/1252, Lahore, Kujah and Sodra, as we saw, were all subject to them, and Jajner

is described as border territory.48 Even Jalal al-Din's defection proved only a temporary setback, since

Juzjani's phrasing betrays the fact that as a result of Sali Noyan's campaigns Sind lay outside the dominions

of his sovereign Nasir al-Din Mahmud Shah.49 The Mongols, who through their satellites controlled

Binban, the Salt Range and the middle and upper Indus valley, now threatened the heartlands of the Delhi

Sultanate. Describing the events of the late 1250s, Juzjani twice refers to Tabarhindh as 'the frontier'.50 It is

the 'frontier of" Islamic territory, such as the province of Sind, Lahore, and the direction (taraf) of the river

Beah' that the Mongols were attacking by 656/1258; when Nusrat Khan exchanged iqta's with Shir Khan in the following year, he received 'the frontiers (sarhadahd) as far as the River Beah fords'; and a few years

into Balaban's reign the Mongols were crossing the Beah.51

In these circumstances, the tradition (hadith) that the Mongol tide would begin to ebb once it

reached Lahore52 provided cold comfort. By 1260 their dominion showed no sign of contraction. Moreover,

the advent of the Mongols was believed to herald the end of time. 'Awfi saw them as the harbingers of Gog

and Magog.53 And had not several authors transmitted the Prophet's statement that the first sign of the end

of time would be the irruption of the 'Turks'?54 Juzjani's reiterated prayer that the sovereignty of Nasir al-

DIn Mahmud Shah would endure until the Day of Resurrection (td qiydm-i qiydmat)55 is perhaps more than

merely sycophantic hyperbole. For Muslims of his generation, the last things were not far off.56 When

relating the Mongol occupation of Uchch and Multan in 655/1257-8, the chronicler permits his anxiety at one point to seep through the skein of his otherwise matter-of-fact account of these events.57

46 Ibid., II, 22, where Iltutmish grants both places as iqta' (KWJAT, to be corrected from BL ms.,

fol. 204a, which has KWJAH; also tr. 750).

47 Ibid., II, 6 (tr. 726-7).

48 Wassaf, 310; JT, ed. Jahn, Indiengeschichte, Ar. text Taf. 57, Pers. text Taf. 22 (and see n.40

above); Qashani, 185.

49 TN, II, 86 (tr. 860). ' 50 Ibid., II, 44 (tr. 793, 794).

51 Ibid., II, 43, 79 (tr. 788, 850-1). TFS, 81. 52 TN, II, 166 (tr. 1136-42).

53 JH, BL ms. Or. 4392, fol. 127b, muqaddima-yi Ya'juj-u Ma'juj.

54 TN, II, 92-4, 98 (omitted in Raverty's tr.).

55 Ibid., I, 422, 450, 462, and II, 189. 56 See also ibid., I, 440, and II, 48 (tr. 597, 800).

57 Ibid., II, 40 (tr. 786). For a useful analysis of Juzjani's view of the Mongols, see D. O.

~ It comes as something of a surprise, therefore, to find the Mongols in turn adopting a more conciliatory stance, in the last contacts recorded by Juzjanl. In response to an indirect approach from Nasir

al-DIn Muhammad b. Hasan Qarluq, who ruled Binban as a Mongol satellite, Ulugh Khan Balaban had sent

a chamberlain (hdjib) with his consent to a marriage alliance between their two families. The envoy's

mission became known to Kiishlii Khan, who alerted the Mongols, and Nasir al-Din Muhammad had to

pass him on to Hiilegii's court in Persia; but he did so, allegedly, with additional letters drafted by himself

but purporting to come from Ulugh Khan. Hiilegii welcomed the hajib and sent him back with his own

emissaries, who were received by the sultan and Ulugh Khan in Rabi' II 658/March 1260. Juzjanl makes

great play of the review outside Delhi in which the Mongol envoys were treated to an impressive

demonstration of the Sultanate's military strength. Hulegu allegedly instructed Sali Noyan that if a Mongol

horse entered Mahmud Shah's dominions its hooves were to be lopped off.58

From what Juzjanl says, Nasir al-DIn Muhammad aimed to pass off Ulugh Khan's response as a

gesture of submission to the Mongols and thereby acquire credit with them as the intermediary. But the

chronology of the Mongol embassy may have a significance which was not apparent to Juzjanl. Hulegu's

representatives had already reached the vicinity of Delhi when Ulugh Khan Balaban left for a brief

campaign against the Meos (Miwat) in Safar 658/February 1260.59 By the time of their despatch (towards

the end of 657/1259) Hiilegii would already have heard of the death of his brother the qaghan. Anticipating

a disputed succession, he may have patched up peace with the Sultanate in order to leave his hands free

while he completed the conquest of Iraq and Syria. Alternatively, the fact that Berke too was in diplomatic

contact with Delhi in this same year has prompted the suggestion that both rulers were actuated by their

rivalry, the one seeking the support of a fellow-Muslim in order to encircle the Ilkhanate, the other the

sultan's neutrality.60 It is equally possible, of course, that Juzjani's story masks a genuine offer of submission by Ulugh Khan, designed to buy time for the Sultanate. Such a scenario would better explain

the emphatic manner in which Sali was forbidden to encroach on Delhi territory, reminiscent of the

privilege of inviolability accorded, for instance, to Lesser Armenia when its king became a Mongol client

in 1254.61 But that the Mongols did not crown their spectacular advance of the

Morgan, 'Persian historians and the Mongols', in Morgan (ed.), Medieval historical writing, 111-

13.

58 TN, II, 83-8 (tr. 856-63).

59 Ibid., II, 79 (tr. 851). 60 Aziz Ahmad, 'Mongol pressure in an alien land', CAJ 6 (1961), 183-4, 185. Aubin,

'L'ethnogenese', 81. See TN, II, 218 (tr. 1292), for Berke's embassy to Delhi.

61 Kirakos Ganjakeci, Patmufyun Hayoc, tr. Robert Bedrosian, Kirakos Ganjakets'i's History of the

Armenians (New York, 1986), 304; see also J. A. Boyle, 'The journey of Het'um I, king of Little Armenia,

to the court of the Great Khan Mongke', CAJ 9 (1964), 181.

~ past decade with the conquest of further Delhi 'territory was due primarily to the outbreak of

strife within the imperial dynasty in 1260.

The advent of the Neguderis

In the Indian borderlands, special circumstances robbed Hulegu of his ability to direct military

operations against the Delhi Sultanate. Some of the Jochid troops in Persia escaped massacre at the hands

of Hulegu's forces and took refuge in Syria and Egypt with the Mamluk Sultan.62 Others fled into present-

day Afghanistan to join Neguder, who commanded the Jochid contingent there. But Hulegu's forces in turn

defeated Negiider's army, which moved eastwards and, according to Rashid al-Din, overran the territory

'from the mountains of Ghazna and Bini-yi Gaw to Multan and Lahore'.63 This is the first mention of Bini-

yi Gaw, a locality which is closely associated with the Negiideris and which is known to have lain not far

from Shal (the modern Quetta).64

Iranian sources, regrettably, have no more to say about the arrival of the Negiideris, or Qara'unas as they were also known (appendix III), but a garbled story picked up in Kirman by Marco Polo a few years

later may throw some light upon it. He makes Negiider's band pass through Badakh-shan, the Pashai and

Kashmir until they reached 'the city Dilivar', which they wrested from its ruler, 'Asidin Soldan', and which

allegedly became Negiider's base.65

The Polo account is here highly confused, and has attracted the

attention of successive commentators.66 There can be little doubt that his 'Asidin' is 'Izz al-Din Kushlu

Khan,67 though in this case 'Dilivar', which has been identified plausibly with Lahore ('citta di Livar'),

presents some difficulty,68 since Kiishlii Khan, who ruled in Sind, is not

62

Jackson, 'Dissolution', 232-3. Ayalon, 'Wafidiya'.

63 Sayfi, 270-2, with the date 660/1262, though in view of Rashid al-Din's chronology 661/ 1263 is

more likely. JT, II, 139 (tr. Boyle, 123/tr. Verkhovskii, 82); and cf. Aubin, 'L'ethnogenese', 80-1, for an

elucidation of the text.

64 Sayfi, 270. For Shal as the alternative name of Quetta, see IG, XXI, 13, 20. The identification of

Bini-yi Gaw with Shashgaw (15 m. N.E. of Ghazna on the road to Kabul), cited in Boyle, 'Mongol

commanders', 247 n.74, is therefore to be discarded.

65 Marco Polo, Le divisament dou monde, tr. A. C. Moule and Paul Pelliot, The description of the

world (London, 1938, 2 vols.), I, 121; tr. Sir Henry Yule, The book of Ser Marco Polo the Venetian, new

edn. H. Cordier (London, 1903-20, 3 vols.), I, 98.

66 Yule, ibid., I, 103 n., rightly observed that Polo conflates two quite distinct episodes, one

involving the Jochid general and the other, which occurred several years later, a Chagha-dayid prince

(actually named Tegiider) who was active in the Caucasus region. See also Pelliot, Notes on Marco Polo,

190-6. The confusion between Negiider and Teguder is repeated in Spuler, Mongolen in Iran, 62, and

Wink, Al-Hind, II, 206, 208. Sir Aurel Stein, 'Marco Polo's account of a Mongol inroad into Kashmir',

Geographical Journal 54 (1919), 92-103, stands in need of revision.

67 Pelliot, Notes on Marco Polo, 52, while agreeing with 'the general opinion that this must be

Balaban' (Asidin > Ghiyath al-DIn), still had doubts and observed that the name 'looks more like 'Izz al-

DIn'.

68 Yule, I, 104-5 n. Pelliot, Notes on Marco Polo, 626, on the other hand, was still prepared to

~ known ever to have held that city; the history of the Lahore region between its abandonment by

Jalal al-Din Mas'ud and its restoration by Balaban in c. 666/1268 is a blank. The only other author to speak

of Kushlu Khan's fate is Isami, who alleges that Kushlu Khan lost Multan to Balaban and was obliged to

take up residence in Binban, though he brought the Mongols into Sind on two subsequent occasions.69 On

the other hand, contemporary poets praise Kushlu Khan's son, Nasir al-Din Muhammad, who was clearly

no hapless exile but a prince of some standing who ruled Uchch and Multan for a few years.70 The whole

question of Kushlu Khan and his dynasty is doubtless destined to remain unresolved.

Simultaneously with these upheavals, two other figures who had played a leading role in events on

the frontier over the previous decade were likewise eliminated. Alughu's forces, at the point when that prince was still aligned with Arigh Boke, arrested Sali and took him as a prisoner to Transoxiana.71 The

downfall of Nasir al-Din Muhammad b. Hasan Qarluq is attributed in the Ilkhanid sources to the intrigues

of Shams al-Din Muhammad Kart of Herat and an otherwise unknown 'Khudawandzada Barghundi; he was

summoned to Hiilegii's court, along with certain other local rulers, and executed on a charge of disloyalty.

A consequence of his removal was the flight to India of the Khalaj leader and future Delhi Sultan, Jalal al-

Din, who had been in his service.72

The Mongols and India after 664/1266

Given what we know of the era of Nasir al-Din Mahmud Shah, it is surprising that some decades

later, in the sketch of the Sultanate's history with which he prefaces his Diwal Rani, Amir Khusraw dated the onset of Mongol inroads from Balaban's reign.73 But if Barani is to be believed, these years did witness

a revival of Mongol pressure on the Panjab. He blames it on the fact that the successors of Shir Khan, who

had held the iqta's in the path of the Mongol advance and had been poisoned on

envisage Delhi; though ibid., 195-6, both 'Deli' and 'Malabar' of Ramusio's version are regarded

with suspicion.

69 FS, 154-5 (tr. 278-80).

70 Siddiqi, 'Historical information', 66. 'Amid Sunnami, cited by Bada'uni, Muntakhab al-

Tawarikh, ed. M. A. 'Ali (Calcutta, 1864-9, 3 vols.), I, 110, 121. See also the laconic mention in TFS, 66.

71 Wassaf, 12. Rashid al-Din mentions only that Alughu sent troops to Samarqand, Bukhara and

other parts of Transoxiana, though adding, confusingly, that he attacked the posses-sions of Berke (like

Alughu, a supporter of Arigh Boke): JT, II, 403-4 (tr. Boyle, 257-8/tr. Verkhovskii, 163). See Barthold,

Turkestan, 488.

72 Wassaf, 311. JT, ed. Jahn, Indiengeschichte, Ar. text Taf. 57, Pers. text Taf. 23 (German tr. 48-9

has 'Mamluken' in error for muluk). Qashani, 186. Cf. also Siddiqui, 'Qarlugh kingdom', 85-6.

73 DR, 50.

~ Balaban's orders, did not share his capacity.74 Even if this is true, the Mongol inroads may

simply reflect the fact that the Neguderis were now launching raids from bases closer at hand than those

occupied by Sali Noyan's army. They almost certainly mounted the attack on Uchch and Multan which a

later tradition placed in the year of the death of the Chishti shaykh Farid al-Din Ganj-shikar (664/1265-6).75

At any rate, annual attacks once more became the pattern.76 Later, Barani asserts that the Mongols regularly

advanced as far as Rupar on the upper Sutlej.77 The restoration of Lahore was the only recovery Balaban

was able to make in the north; but there is, significantly, no mention of its being granted as iqta' again prior

to the fourteenth century.

Sultan Balaban entrusted the task of guarding the western frontier to his sons: the elder,

Muhammad, was granted Sind, and the younger, Bughra Khan Mahmud, was stationed at Samana. The two princes shared responsi-bility for defence against the Mongols with the barbeg, Ikhtiyar al-Din Begbars,

and BaranI pays tribute to the effectiveness of these arrange-ments.78 Around 680/1281-2, however, Bughra

Khan was transferred permanently to Lakhnawti. It is possible that this weakened the frontier defences in

the last years of Balaban's reign, since 'IsamI records an invasion by two bands of Mongols in which the

force sent to repel them by Muhammad suffered a reverse.79 A heavier blow was to fall in the winter of

683/1284-5, when Muhammad himself was defeated and killed in battle with the Mongol commander

Temur.80 According to the most circumstan-tial account of the engagement, given in a marthiya by Amir

Hasan Dihlawi which is preserved by Sirhindi, it took place at Bagh-i *NIr, close to the junction of the

Ravi and the Greater Dhandh, on 29 Dhu'l-Hijja 683/8 March 1285. Amir Hasan's friend and fellow-poet,

Amir Khusraw, who was briefly taken prisoner by the Mongols, commemorated the disaster in his Wasat

al-Hayat.81

Temur again invaded India and ravaged the territory between Lahore

74 TFS 65-6. Barani's analysis contradicts what we know about the disposition of iqta's at the time

Juzjani wrote, when Shir Khan had been moved from Tabarhindh to Bhayana (see above, p. 112).

75 Amir Hasan Dihlawl, Fawaid al-Fu'ad, ed. M. Latif Malik (Lahore, 1386/1966), 373-4.

76 WH, IOL Persian ms. 412, fols. 90a, 134b. TFS, 82; and see also 50-1, where a speech put into

Balaban's mouth refers to annual invasions.

77 Ibid., 82, with the corrupt reading 'ZWBR: text restored by Hodivala, Studies, II, 85-6. On Rupar,

'a town of considerable antiquity' situated at 30° 58' N., 76° 32' E., see IG, XXI, 339.

78 TFS, 80, 81: for the bdrbeg's name, see above, p. 78.

79 FS, 171-3 (tr. 299-300).

80 The description of Temiir as 'one of the great Chingizi amirs, to whom belonged Herat,

Qandahar, Balkh, Badakhshan, Ghaznayn, Ghur and Bamiyan' is not found in any source earlier than

Firishta (I, 143) and is consequently suspect; it is nevertheless accepted by Aziz Ahmad, Political history,

285. 81 Quoted at length by Bada'uni, I, 138-55 (extensive citations in Mirza, Life and works, 56-9). See

also TFS, 109 (with the year 684); FS, 175-81 (tr. 304-11). The date is supplied by Amir Hasan (quoted in

TMS, 45) and by Khusraw in WH, IOL ms. 412, fols. 133a, 134b

~ and Samana early in 686/1287, when the new sultan, Balaban's grandson Mu'izz al-Din

Kayqubad, had begun to move east in preparation for the confrontation with his father Bughra Khan in

Awadh. On the approach of the barbeg Khan Jahan Shahik, the Mongols retreated without offering battle

and, according to Sirhindi, were pursued as far as the foothills of Jammu.82 But with the decay of the

sultan's authority, and the concentra-tion of power in the hands of Nizam al-DIn (above, pp. 53, 81), many

nobles were eliminated, among them Shahik (now Azhdar Khan and amir of Multan).83 Responsibility for

frontier defence appears thereafter to have fallen principally to the Khalaj amir Jalal al-Din Firuz, whom Balaban had made muqta' of Kaithal and na'ib of Samana.84 But although Jalal al-Din is portrayed as a

veteran of the Mongol front by the time he ascended the throne in 689/1290, we know virtually nothing of

his exploits. Amir Khusraw simply puts into the new sultan's mouth references to campaigns against the

Mongols of 'Ghaznayn, Kurraman and *Birjand', and later alludes to his intention of advancing from

Multan towards Ghazna; Barani is even less specific.85

Soon after his accession, the sultan, who had entrusted the iqta' of Multan to his second son, Erkli

Khan, found time, amid the victories of a twelve-month period commemorated in Amir Khusraw's Miftah

al-Futuh, to march against the Mongols. This campaign, from which he returned to Delhi at the onset of

690/1291 after an absence of one month, was directed against a region that cannot be identified.86 It

apparently provoked the next Mongol assault, in 691/1292, headed by the prince 'Abd-Allah. The sultan made camp in a locality called by both BaranI and 'IsamI 'Bar Ram', where a river (the Sutlej?) separated

the two armies. After some skirmishing between the two vanguards, however, a truce was declared. Jalal

al-DIn and 'Abd-Allah exchanged friendly messages and gifts, and the Mongol prince withdrew, leaving

behind under a commander named Alughu a group of his followers who accepted Islam and who were

settled by the sultan in the neighbourhood of Delhi.87 Thereafter we hear of no further invasions during

Jalal al-Din's reign. Following the old sultan's murder in 695/1296 by his nephew 'Ala' al-Din, his sons,

having held out for a time in Multan, were forced to surrender and were later put to death. This crisis may

well

(cited by Bada'uni, I, 147). The location given in WH varies: on the borders of Multan (fol. 78a),

or in the vicinity of Lahore (fol. 132b).

82 QS, 62-5. TMS, 54.

83 TFS, 134.

84 Ibid., 170, 194, 195.

85 MF, 8. DR, 51. TFS, 196. Birjand may be identical with 'Barghund' (above, p. 116), an

alternative name of Naghar: Hodivala, Studies, I, 168; A. D. H. Bivar, 'Naghar and Iryab: two little-known

Islamic sites of the north-west frontier of Afghanistan and Pakistan', Iran 24(1986), 131-8.

86 MF, 22, 23. The region is named as BALAGHTRK: the same reading is found in IOL Persian mss.

51, fol. 490a (margin), and 412, fol. 788b (margin).

87 TFS, 218-19. A different account of this episode is furnished in FS, 209-14 (tr. 372-9), where

the Mongols retreat after an indecisive battle with Jalal al-DIn's brother, Malik Khamush.

~ have weakened the defences of Sind, a situation of which, as we shall see in chapter 11, the

Mongols were not slow to take advantage.

The Negiideris and their Mongol neighbours The collapse of Mongol unity after Mongke's death had deposited in the Indian borderlands a body

of Mongol troops with no allegiance to either the Ilkhans or the Chaghadayids; the region was now the

camping-grounds of a smaller, independent grouping without access to the resources of the whole Mongol

empire. The strength of the Negiideris was sufficiently modest not to jeopardize the survival of the

Sultanate, but still powerful enough to form yet another barrier in India's defence, obstructing any

expansionist tendencies on the part of the Mongols of Persia or of Transoxiana. Marco Polo heard tell of

Neguder that he 'makes war on all the Tartars who dwell round about his kingdom'; and certainly the

Negiideris acquired a name for brigandage and highway robbery and were notoriously unready to submit to

any ruler.88

Their arrival in the Indian borderlands drove a wedge between India and Hulegu's dominions.89

Ilkhanid sources speak of Negiideri raids on Fars and Kirman and assert that the people of Fars lived in fear of such raids down until the end of the reign of the Ilkhan Arghun (d. 690/1291).90 When the Chaghadayid

khan Baraq invaded Khurasan in 668/1269-70, Hulegu's son and successor, Abaqa, sought to deflect him

with the offer of Ghazna and 'Kurraman-i Binban'91 - regions that were currently not in his gift. The Ilkhans

may not have been primarily interested in this tract, and may at this stage have envisaged leaving its

reduction to the Chaghadayids. This is not to say that they made no effort to exert indirect influence over

the Negiideris or to mount punitive expeditions against them. Both Hulegii and Abaqa despatched a series

of commanders against Neguder and his forces.92 During Baraq's attack on Khurasan, various Chaghadayid

princes deserted to the Ilkhan, and Abaqa sent one of them, Mubarak Shah, to head 'the army of Negiider in

the confines of Ghaznayn'.93 But Mubarak Shah's death in 674/1275-6 while leading an attack on the

province of Kirman suggests that his allegiance to the Ilkhan was superficial.94 Abaqa had also

88 Marco Polo, tr. Moule and Pelliot, I, 122/tr. Yule and Cordier, I, 99. Sayfi, 432.

89 Jahn's view ('Zum Problem', 618) that it was the constant need to defend their other frontiers

which prevented Hulegii's successors from making good their claims to present-day Afghanistan fails to

take account of the situation that had emerged there.

90 JT, III, 151-2 (tr. Arends, 94). Ibn Zarktub, Shiraz-Nama (c. 1344), ed. Bahman Karimi (Tehran,

1310 Sh./1932), 66/ed. Ismail Wa'iz Jawadi (Tehran, 1350 Sh./1971), 91-2. Wassaf, 199-203.

91 JT, III, 122 (tr. Arends, 78); and cf. Sayfi, 308.

92 Aubin, 'L'ethnogenese', 80 n.3, for textual references. 93 JT, II, 193 (tr. Boyle, 153-4/tr. Verkhovskii, 100).

94 Aubin, 'L'ethnogenese', 83. Anonymous, Ta'rlkh-i Shahi-yi Qarakhita iyyan (late thirteenth

century), ed. Muhammad Bastani-Parizi (Tehran, 2535 Shahanshahi/1976), 248-50.

~ Map 2: The frontier with the Mongols

~ entrusted another Chaghadayid, Bojei, with a command among the Negu-deris, and Bojei's son

'Abd-Allah, who would later invade India (above, p. 118), appears to have succeeded him by this

juncture.95 In 678/1279, in reprisal for a Neguderi raid on Kirman and Fars, Abaqa himself led an army as far as Herat, receiving the submission of Mubarak Shah's sons, while his son Arghun was sent on ahead

into Ghur and Gharchistan. A number of Negiideri amirs and their dependants were subsequently employed

in the Ilkhans' service in western Persia.96

These Ilkhanid campaigns, it seems, had strictly limited aims. The hypothesis advanced by

Professor U. N. Day, that the sudden spate of Mongol incursions into India from c. 1285 reflects the

disorders in Persia after Abaqa's death in 680/1282,97 is difficult to sustain. The Mongols of Persia were cut

off from India not only by the Negiideris but also by the sometimes recalcitrant Kartid kingdom of Herat

and by the virtually autonomous kingdom of SIstan. Abaqa's successor Ahmad included Tiginabad in a

grant to the malik of Sistan in 683/1284, and indeed SIstan is described as still not subject (7/) to the Ilkhan in the reign of Ghazan (694-703/1295-1304).98 The most we can say is that the Ilkhans' operations in

eastern Khurasan may have encouraged the Negiideris to devote more attention to the Panjab. But there

were certainly other, more local stimuli at work. The seizure of Qandahar in 680/1281 by the Kartid malik

of Herat, Shams al-Din Muhammad II,99 would surely have menaced the Negiideris, and may therefore help

to explain an apparent increase in Mongol pressure on the Delhi Sultanate in Balaban's last years.

As for the Chaghadayids, Alughu's advance into the Indian border

95 Kirmani, Simt, 49. JT, II, 177, hakim-i charik-i Qarauna bud dar hudud-i Ghaznayn (tr. Boyle,

144; not in text used for Verkhovskii's tr.)- Anonymous (fifteenth-century) Chinggisid genealogy, Mu'izz

al-Ansab, BN ms. Ancien fonds persan 67, fol. 29b, for Bojei's flight to Abaqa. In JT (but not in his

genealogical work, SP, TSM ms. HI Ahmet 2937, fol. 117b), Rashid al-Din seems to list this branch of the Chaghadayids twice, and we need not assume, with Aubin ('L'ethnogenese', 84 n.l), that there were two

princes called Bojei. Barani calls ‘Abd-Allah 'grandson of Hulu [i.e. Hulegu]' (TFS, 218), an error which

misled Lal, History of the Khaljis, 30; but it possibly reflects awareness at Delhi that 'Abd-Allah had been

an Ilkhanid appointee.

96 JT, III, 152-3, 252 (tr. Arends, 94, 143); cf. also JT, II, tr. Verkhovskii, 100 (sentence omitted

from Blochet's text; see Boyle tr., 154 n.40). On this campaign, see Aubin, 'L'ethnogenese', 85-6; and for

the subsequent history of the Negiideri contingents trans-ported westwards, ibid., 87-90.

97 U. N. Day, 'The north-west frontier of the Sultanate', in his Some aspects of medieval Indian

history (New Delhi, 1971), 43-4; and cf. also his 'The north-west frontier under the Khalji Sultans of Delhi', IC 39 (1963), 99.

98 Jacfar b. Muhammad b. Hasan Ja'fari, Ta'rikh-i Yazd, ed. Iraj Afshar (Tehran, 1338 Sh./ 1959),

27; for the grant, see Bosworth, History of the Saffarids, 435. On the kingdom of Herat, see T. W. Haig and

B. Spuler, 'Kart', Enc.Isl2; also Spuler, Mongolen in Iran, 129-33, and his 'Das heutige afghanische Gebiet

und sein Ringen urn Selbstandigkeit gegeniiber die Mongolen im 13./14. JH.', in G. L. Ulmen (ed.), Society

and history. Essays in honor of Karl August Wittfogel (The Hague, 1978), 403-9.

99 Sayfi, 369-73: the date, of course, is not above question (above, p. Ill, n.39).

~ regions - instanced in the overthrow of Sali Noyan - appears to have been a temporary

phenomenon. Despite Aubin's assertion that 'the impossibility of maintaining contact with the Golden Horde soon condemned the Neguderis to a change of masters which was already complete around 1270',

we may well question whether the Chaghadayids did effectively assert their authority here at this time.100

Still less could they have retained it in the troubled years that followed. What links, if any, the renegade

Chaghadayid princes whom the Ilkhans installed in the Ghazna region retained with the Chaghadayid ulus

is not clear. In his account of Chaghadai's line, Rashid al-Din, writing at the very beginning of the

fourteenth century, speaks of 'the province of Ghazna and the Qara'una army, which has long had

connections with them'.101 But perhaps we should not read too much into this. The political organization of

the Negiideris was doubtless a loose one, and the influence of the head of Chaghadai's ulus - like that of his

enemy, the Ilkhan - varied at different times and from one contingent to another.102 Not until the last decade

or so of the thirteenth century was a successful effort made from Transoxiana to bring the Negiideris under

control and to dominate the marches of India (see chapter 11); and then the impact on the Delhi Sultanate would be felt more keenly than at any previous time.

100 Cf. Aubin, 'L'ethnogenese', 82: 'L'impossibilite de maintenir le contact avec la Horde d'Or

condamna bientot les Neguderi a un transfert de sujetion qui etait deja chose faite vers 1270. Les

Cagataides en furent les beneficiaires, et non les Ilkhans.'

101 JT, II, 174, lashgar-i Qara'una ki az qadim bdz ba-ishan ta'alluq dashta and (tr. Boyle, 142/tr.

Verkhovskii, 92,-3).

102 See the remarks in Aubin, 'L'ethnogenese', 87.

~ CHAPTER 7

Raid, conquest and settlement

To chart the progress of Muslim arms during the thirteenth century is by no means an easy task.

There are no contemporary Hindu narrative sources, properly speaking; even the epic Hammiramahakavya,

for example, dates from the end of the fifteenth century. From the Hindu side, inscriptions -whether on

stone or in the form of copper-plate grants issued by rulers to their subordinates - are the best evidence we

possess, although the references there to Mlecchas ('unclean ones'), Turushkas ('Turks') and Yavanas

('Westerners') are frequently vague. Muslim writers, less than forthright about reverses, provide

fragmentary data. To notice, moreover, that Juzjani's accounts of military operations are largely confined to those on which Nasir al-Din Mahmud Shah or Ulugh Khan Balaban were present is to recognize the extent

of our ignorance. There must have been numerous military operations conducted at a local level by amirs

and muqta's which won fresh territory for the revenue-collector and the settler, but of which we know

nothing. Although the distribution of iqta's may serve as a pointer to Muslim conquest, the information

available for any given time is hardly exhaustive. Here and there an inscription dating the construction of a

mosque confirms the presence of a Muslim community; but even so we cannot tell whether this was the

earliest mosque to be built in the town concerned.

It is important not to be misled by the terminology of holy war (jihad, ghazw) employed by

Juzjani and others. The primary aims of the Islamic holy war are the defence of the Dar al-Islam and the

extension of Muslim rule over pagan territory. In the circumstances of the early Sultanate period, the latter aim could only be realized in certain regions for perhaps a limited time; it might not be realizable at all. A

case in point is the great fortress of Ranthanbor in Rajasthan, which was taken twice, and repeatedly

attacked, before its final capture by 'Ala' al-Din Khalji in 700/1301. During the thirteenth century the pious

Muslim monarch might have to be content with swashbuckling raids. In the more spectacular cases, the

capital of a Hindu kingdom was taken, looted and then abandoned so that its ruler was able to reoccupy it in

the wake of the Muslim army's departure. Assaults of this

123

~ kind rallied the faithful, weakened an infidel prince by depriving him of treasure, horses and

elephants and diminished his standing in the eyes of his peers and his subjects. But many expeditions

would have been designed simply to replenish stocks of cattle and slaves. For much of the period after 633/1236 Muslim domination either remained static or receded as the Sultanate proved unable even to hold

on to the acquisitions made by Mu'izz al-Din and Aybeg or by Iltutmish.

Predatory incursions into Hindu territory might also have other main-springs than conventional

religious fervour. Ulugh Khan Balaban is said to have recommended military activity against the Hindu

powers to Nasir al-Din Mahmud Shah in order that plunder wrested from the Hindus might be used to pay

the troops that resisted the Mongols.1 Such campaigns also often served the interests of the individual

commander. Balaban's own lucrative expedition to Ranthanbor in 652/1254, which immediately pre-ceded

his return to power at court; Toghril's profitable raid on eastern-Bengal just prior to his insurrection in c.

678/1279-80; and still more the booty obtained from the Deccan by 'Ala' al-Din Khalji in 695/1296 which

greased his path to the throne, suffice to remind us of that.

Strongholds and refuges

Earlier we saw how the Ghurid conquests established a basis for Muslim rule in the north

Gangetic plain, while leaving certain Hindu rulers on their thrones in return for the payment of tribute. At a

local level, power remained in the hands of a host of lesser chieftains who frequently defied the sultan's

government. When he wanted to conjure up a vivid image of the peace and order that prevailed under 'Ala'

al-DIn Khalji, it was enough for BaranI to depict these chiefs (muqaddams) and headmen (khutslkhots) as

standing guard on the highways and keeping watch over travellers and caravans.2 Ibn Battuta, who visited

the Sultanate during Muhammad b. Tughluq's reign, was careful to distinguish Hindus who lived in villages subject to a Muslim officer (hakim) from those he terms 'rebels and warriors who maintain themselves in

the fastnesses of the mountains and plunder travellers'.3 And he returns to the theme at a later juncture:

The infidels in the land of India inhabit a territory which is not geographically separated from that

of the Muslims, and their lands are contiguous, but though the Muslims have the upper hand over them yet

the infidels maintain themselves in inaccessible mountains and rugged places, and they have forests of

reeds ... The infidels live in these forests, which for them are as good as city walls, and inside them they

have their cattle and grain and supplies of water collected from the rains, so

1 TN,II,57(tr.816).

2 TFS, 324; and see also 340. For the term khot, see Yule and Burnell, Hobson-Jobson, 480-1; S.

H. Hodivala, 'Notes on Hobson-Jobson', IA 58 (1929), 173; idem, Studies, I, 277-8.

3 IB, III, 133 (tr. Gibb, 612).

~ that they cannot be overcome except by strong armies of men who go into these forests and cut

down those reeds.. .4

These are the mawasdt (sing, mawds, 'shelter', 'refuge') that figure so frequently in our Indian

Muslim sources. The word applied to any of the countless regions of broken terrain, arduous defiles and

jungles where the Muslim heavy cavalry could barely penetrate and the enemy could hold out with relative

impunity.5 One such region was surely the tarai beyond the river Saru (Sarju) - 'the abundant jungles of Hindustan, the narrow passes and the torrents, and the dense foliage of numerous trees', as Juzjani puts it -

where Balaban advanced in 654/1256 in the fruitless pursuit of his Muslim enemies.6 The mawds, of

course, served as a refuge also for Muslim rebels like Qutlugh Khan, who passed through such territory on

his way to seek asylum in Santur in 654/1256;7 although sometimes their hopes were cheated, as when

Malik Chhajju in 689/1290 and the adherents of 'Ayn al-Mulk Ibn Mahru in c. 740/1339-40 were handed

over to the sultan after seeking asylum in a mawds.8 The term grew to be synonymous with defiance. 'All

Gawr and Tirhut became mawas', says 'Isami when referring to the Muslim-led secession of Bengal and

Bihar in Muhammad b. Tughluq's reign.9

Hindu warriors were often ready to side with Muslim rebels against the government. Toghril's

insurrection in c. 678/1279-80 drew in large numbers of foot from the landed gentry of Bengal - those

'renowned paiks', as Barani calls them, who subsequently paid a heavy price when the victorious sultan Balaban had them all beheaded.10 No more successful were the rawats and paiks - termed merely 'some

infidel troops from Hindustan' (az Hindustan sipahi-yi chand bi-din) by Amir Khusraw - who gathered

around Balaban's nephew Malik Chhajju at Kara in 689/1290 in an attempt to overthrow the Khalji regime.

They paraded before Chhajju, taking up the betel-leaf (tanbul) in their familiar ceremony of pledging

allegiance, and boasted how they would fall upon Sultan Jalal al-DIn's canopy (chatr) in the heat of battle;

but instead they were rounded up and taken prisoner when Chhajju lost his nerve and fled.11 Kara appears

to have been a veritable nursery of such auxiliaries for Muslim rebels. Not only did 'Ala' al-Din recruit two

thousand paiks here only a few years later, but we find 'drug-quaffing paiks' allegedly behind a rising in c.

1338 in the same locality.12

4 Ibid., III, 389 (tr. Gibb, 741-2). 5 Hodivala, Studies, I, 226-9. Examples are found in TN, I, 491, and II, 17-18, 19, 26, 27, 29,

47,52,53,57,61,72,76-7.

6 Ibid., II, 71, jangalha-yi Hindustan-i gashn wa-mada'iq-i lurha wa-iltifaf-i ashjar-i bisyar

(Raverty's tr., 837-8, modified).

7 Ibid., II, 72 (tr. 839); cf. also I, 491 (tr. 704-5 inaccurate).

8 TFS, 182, 490-1.

9 FS, 606 (tr. 902).

10TFS, 83, 91.

11 Ibid., 182-3. For Amir Khusraw's dismissive phrase, see MF, 8.

12 TFS, 222, 487. For the reservoir of armed men on which the Mughals later drew, see Dirk

~ In describing the Muslim and Hindu territories as geographically unsepa-rated, Ibn Battuta was

pointing to a feature of the Sultanate that marked it out from other Muslim polities. Elsewhere in the

Islamic world it made sense to talk of the Dar al-Islam and the Dar al-Harb; but not in India. A series of dots would indicate the extent of the sultans' rule with greater realism than does the uniform shading

favoured in historical atlases. Obeisance (paibus, zaminbus), like tribute, was intermittent. Barani makes

Nizam al-Din advise Sultan Kayqubad to advance to meet his father in 686/ 1287 at the head of an

imposing army, in order that the rais and ranas might be induced to wait upon him en route.13 The open

countryside, the forests, the hills - these were the domain of the infidel. The Muslim population of the

Sultanate largely resided in its fortified towns and cities, and even there they were not unusually a minority.

Ibn Battuta observes of the great fortress of Gwaliyor that it was 'an isolated and inaccessible castle, in the

midst of the infidel Hindus', and that the generality of its inhabitants were infidels; so too were the majority

of the people of the neighbouring town of cAlapur.14 Not long afterwards we find 'Ayn al-Mulk Ibn Mahru,

as governor of Sind for the Tughluqid Firtuz Shah, commenting acidly on 'peasants (dahaqin) and

landholders (zamindars) who are only ostensibly subjects (ra'aya-yi suri) and pay tax out of fear of the army or the blow of the sword'.15 From each strongpoint the authority of the Muslim governor and tax-

collector radiated outwards for a distance that waxed and waned with the conduct of local military

operations or the proximity of a large field army sent from Delhi. In India the 'war zone', peopled by harbis,

was never far away.

The heartlands

We are told very little of the relations of the sultan's governors in Sind with local Hindu powers.

Sultan Balaban's son Muhammad, when governor of Multan, was married to the daughter of a Hindu rai,

who ransomed his son-in-law's body from the Mongols in 684/1285.16 When not intent on repelling

Mongol attacks, Muhammad seems to have been engaged principally in asserting his authority over the

partially Islamized Sumra princes of the Indus delta. Having been tributary to Qubacha, they had submitted to Iltutmish, and Diwal (Daybul) and Damrila are said to have been subject to

H. A. Kolff, Naukar, Rajput and Sepoy: the ethnohistory of the military labour market in

Hindustan, 1450-1850 (Cambridge, 1990), chap. 1, esp. 3-17.

13 TFS, 141; cf. also 108, where the chiefs come to wait upon Balaban after the Bengal campaign.

14 IB, III, 188, 195, and IV, 29 (tr. Gibb, 642, 645, 785).

15 IM, 75. For the varied groups who made up the class loosely termed zamindars, see I. Habib,

'Agrarian economy', 58-9. 16 FS, 180-1 (tr. 311).

~ Radiyya.17

But their allegiance could not be taken for granted, and it must have fallen to the

sultan's representatives at Siwistan to keep watch on them. Muhammad's iqta' is said to have extended as

far as Janani, a town some 120 miles upstream from Thatta and known to have been held by the Sumras;

but we know that in c. 680/1281-2 he reduced the Sumra strong-hold at Damrila.18 At some point early in

the fourteenth century, says 'Afif, 'Ala' al-Din Khalji mounted an unsuccessful campaign in the region,

although we also know that his muqta' at Deopalpur, Ghazi Malik Tughluq, constructed in lower Sind a

new fortress, which he named Ghazipur.19 Thereafter, no more is heard of this territory until the reign of Muhammad b. Tughluq.

The Panjab was home to a number of imperfectly subdued tribes, notably the Khokhars, whose

original territory lay between the Jhelam and the Chenab but who were by now encroaching on the regions

east of the Beah.20 They chiefly threatened Lahore. In 639/1241 Qaraqush Khan, the muqta' of Lahore,

massacred a band of Khokhars whom he found scaven-ging in the stricken city for anything the Mongols

might have left behind.21 According to Barani, Balaban's cousin Shir Khan subdued the Khokhars and other

tribes, a feat which proved beyond the capacity of his successors; and Amir Khusraw credits Jalal al-DIn

Khalji too with harrying them.22 But in the following century we learn from one of Ibn Mahru's letters that

the road from Multan to Ajudhan (now Pakpattan) was regularly harassed by marauding Khokhars.23

Further east, and closer to the heart of the Sultanate, Muslim governors based at fortresses like Tabarhindh, Sunnam and Samana had to contend with other turbulent peoples - the Bhattis and Mains of

the Abuhar region, the Mandahars of Kaithal, and the Jats. They were for the most part hardy pastoralists

who nomadized in the riverine tracts, and Barani, while a prisoner in Bhatner, saw their talwandis, laagers

formed out of wagons within which they gathered their livestock close to a source of water.24 Todars and

Jats, as well as Khokhars, are found among the troops with

17 TN, I, 459 (omitting Damrila, but cf. Raverty's tr. 640). Damrila is Danbharlo, 150 m. E. of

Karachi: J. A. Boyle, 'Jalal al-Din Khwarazm-Shah in the Indus valley', in Khuhro (ed.), Sind through the

centuries, 129 n.13.

18 FS, 175 (tr. 304): immediately prior to his overthrow by the Mongols (683/1285). DGK, 69.

WH, IOL Persian ms. 412, fol. 140b; hence Bada'uni, I, 154 (WMRBLH in error). TMS, 43, for the southern

boundary of Muhammad's iqta'. For Janani, see IB, III, 101 (tr. Gibb, 596-7).

19 Ghazipur: SFS, 91. 'Ala' al-DIn: 'Afif, 251; the late and contradictory testimony on this is

reviewed by Hodivala, Studies, II, 133-4.

20 Abdus Subhan, 'Khokar', Enc.Isl2. For their original territory, see Taj, fol. 190a.

21 TN, II, 165-6 (tr. 1136). 22 TFS, 65. DR, 52. 23 IM, 168.

24 TFS, 568: and cf. Hodivala, Studies, I, 305, for a translation of this passage. On these tribes, see

generally H. A. Rose, A glossary of the tribes and castes of the Punjab and North-West Frontier province (Lahore, 1911-19, 3 vols.), II, 101-6 (Bhattis), 357-77 (Jats), and III, 65 (Mandahars), 102-4 (Minis or

Mains); also Hodivala, Studies, I, 295-6. The Jats were abandoning their pastoralism for agriculture: Irfan

Habib, 'Jatts of Punjab and Sind', in

~ which Radiyya and *Altunapa opposed Bahram Shah in 638/1240.25 The fortress at Bhatner was

built by Shir Khan as muqta' of Deopalpur, and he, according to Barani, reduced these peoples to

obedience.26 But the future sultan Jalal al-Din Khalji fought against the Bhattis and the Khokhars and, when

conducting a raid on Mandahar territory as muqta' of Kaithal (i.e. in the late 1280s), sustained two wounds

in the face that left him scarred ever afterwards.27 And periodic resistance continued into the fourteenth

century, for Muhammad b. Tughluq was to head a punitive campaign against the Mains, the Bhattis and the

Mandahars in c. 1337. It would have been some such group which had attacked Ibn Battuta's party in 734/1334 in the plain between Ajudhan and Abuhar.28

Immediately south of Delhi lay Alwar, a hilly region pitted with defiles and ravines and in the

thirteenth century heavily forested. This, to the Muslims, was the kuhpaya, 'the highlands'. An alternative

name was 'Miwat', after its inhabitants, the Meos, who appear to have been loosely subject to the Chawhan

(Chahamana) kings of Ranthanbor,29 and whose depredations reached across Hariyana in the north and to

Bhayana in the east. Muslim outposts had been created in this tract at Rewari, Narnawl, Palwal and Kama,

which appear as iqta's under Iltutmish and his immediate successors: Rewari was for a time the assignment

of Balaban, who early in his career is said to have reduced to obedience 'the mawasat of the kuhpaya.30

In

658/1260, when his dependants on the outskirts of Hansi had suffered from Meo raids, Balaban led two devastating cam-paigns deep into their territory, bringing back 250 of their leading men for execution in

Delhi and putting thousands of Meos to the sword.31 Juzjani's fulsome narrative might easily persuade us

that the Meos had been suppressed for all time, but for Barani's claim that during the reigns of Iltutmish's

offspring the Meos had continued unchecked, so that they were robbing the mansions (saraiha) in the

neighbourhood of Delhi itself and harassing the water-carriers at the Hawd-i Shamsi. Balaban gave priority

to crushing them, and spent the first year of his reign clearing the jungles in the environs of the capital and

slaughtering the Meos. He constructed a fortress at Gopalgir and established various redoubts (thanaha).

After this, alleges Barani, the citizens of Delhi were spared the

Harbans Singh and N. Gerald Barrier (eds.), Panjab past and present. Essays in honour of Dr.

Ganda Singh (Patiala, 1976), 92-103.

25 FS, 139 (tr. 259).

26 TFS, 65.

27 Ibid., 195. DR, 52.

28 IB, III, 133-5 (tr. Gibb, 612-13). For Muhammad's campaign, see TFS, 483.

29 See generally J. Burton-Page, 'Me'0', Enc.Isl2, and 'Mewat', ibid. For the link with Ranthanbor,

TN, II, 58-9 (tr. 818); Habibullah, Foundation, 152, 153.

30 TN, II, 6, 8, 41 (tr. 726, 730, 790); and for Balaban, II, 52 (tr. 806-7). For Kama, in the

Bharatpur territory, 39 m. N.W. of Mathura, at 27° 40' N., 77° 20' E., see Raverty, 790 n.9; for Palwal, see

also Shu'aib, 'Inscriptions from Palwal', 2-3; RCEA, X, 56, 72-3 (nos. 3678, 3703).

31 TN, II, 78-83, 89 (tr. 850-6, 864); also a brief reference to the first campaign at I, 496 (tr. 715).

~ threat of the Meos.32 The mosque constructed at Narnawl in 671/1272 may indicate the success

of Balaban's policy.33 Under Jalal al-Din Khalji, Narnawl again appears under Muslim rule, and the fact

that on his outward march against Ranthanbor in 690/1291 the sultan was able to move by way of Rewari

and Narnawl suggests that he anticipated no trouble from the Meos.34 But a chance remark by Ibn Battuta

reveals that Muhammad b. Tughluq had been obliged to send troops into the hilly regions near Delhi hot

long before the Moroccan traveller's arrival.35 Nothing throws into sharper relief the limitations of Muslim governmental authority than these recurrent crises in a territory so close to the capital.

The north and west

Beyond the Khokhars lay the people of the Salt Range (Kuh-i Jud). Their suppression by Mu'izz

al-Din in 601/1204 had not cowed them for long, and their conversion to Islam had been merely temporary.

In 643/1245 the raja of the Salt Range, 'Jaspal Sihra', acted as guide to the invading Mongol army, for

which he incurred a punitive attack by the Delhi forces under the army-hajib Balaban in the following

year.36 Soon after his accession as sultan in 664/1266, Balaban led another expedition to the Salt Range,

bringing back the raja's two sons as hostages:37 they appear to have accepted Islam, and are later found

enrolled among the nobility (above, p. 79).

During the thirteenth century the sultans seem to have had little contact with the Hindu princes of

the mountains to the north, for which the term Qarachil is employed by Juzjani and by Ibn Battuta;38

Muhammad b. Tughluq is the first sultan known to have launched a campaign to these distant regions. Our

sources barely mention Kashmir and Jammu, each under its own dynasty of rajas, and there is no evidence

that they ever acknowledged the overlordship of Delhi. But even in Juzjani's time expeditions were sent

into the territories lying due north of Delhi. When the raja of Santur, 'Ranpal, paramount among the

Hindus', gave asylum to

32 TFS, 55-7. For Gopalgir, see also FS, 164 (tr. 291). 33 S. S. Hussain, 'A new inscription of Sultan Balban from the State Museum, Patiala', EIAPS

(1972), 1-3; ARIE (1973-4), 13, 181 (no. D250).

34 MF, 25; ibid, 28, for an amir of Narnawl. For this place, at 28° 3' N., 76° 10' E., see IG, XVIII,

380-1.

35 IB, III, 293 (tr. Gibb, 696-7). 36 TN, II, 56-7 (tr. 815); cf. also I, 479 (tr. 678-9).

37 TMS, 40. TFS, 59-60, mentions the campaign only briefly.

38 TN, II, 126 (tr. 1046); the form is given in scriptio plena by IB, III, 325. The name is Turkish

and appears to mean 'opening in the snow', from qar, 'snow', and achil-, 'to open': Clauson, Etymological

dictionary, 26, 641; Pelliot, Notes sur I'histoire de la Horde d'Or, 64 ('qaracil... endroit qui ne gele pas

dans une surface gelee'). The suggested link with Sanskrit achala, 'mountain', and hence with the location

'Kularchak' (recte Kularchal), near Kashmir, mentioned in the eleventh century by al-Biruni, is therefore

groundless.

~ Balaban's enemy Qutlugh Khan, Balaban sacked his residence at Silmur (Sirmur) in 655/1257.39

Rajasthan and the Siwalik

The northern part of the Aravalli range formed the backbone of the territory called by the Muslims 'Siwalik' (not to be confused with the modern usage, which refers to a section of the sub-Himalaya). This

was a vast area, stretching from Hansi and Sarsati as far south as Nagawr, Ajmer, Sambhar Namak and

Mandor; and indeed a phrase of Barani's suggests that it was deemed to include Jalor also.40 Early in his

reign, Iltutmish led an expedition to Jalor, ruled by a branch of the Chawhan dynasty. Its king, Udayasimha

('Udaysa'), was reduced to submission and undertook to pay tribute.41 The arrangement does not seem to

have been long respected, since Iltutmish in 624/1227 attacked and captured Mandor, which is known to

have belonged to Udayasimha, and indeed an inscription of his son Chachigadeva proclaims that

Udayasimha had 'curbed the pride of the Turushka'.42 Perhaps this is an allusion to the recovery of Mandor,

which figures as part of an iqta' only once, in 639/1242, and was clearly lost by the Muslims at some later

date, since it had to be retaken by Jalal al-Din Khalji in 691/1292.43 Of Jalor, nothing more is heard until

the following century.

Ajmer, the former capital of the main branch of the Chawhan dynasty, had been seized by Aybeg

following its occupation by Hariraja, brother of Prthviraja III. During the reign of Iltutmish it was the

centre of an iqta' which comprised also Lawa, Kasili and Sambhar Namak, and was held for a time by Nasir

al-Din Aytemiir al-Baha'i. But these latter districts do not appear again as Muslim territory in the thirteenth

century, and when we next find Ajmer granted as iqta' it is in conjunction with Mandor and Nagawr, as part

of the large assignment entrusted to Kiishlu Khan in 639/ 1242.44 By this time the southernmost regions of

the Siwalik would have been under heavy pressure from the Chawhans at Ranthanbor, and the grant

perhaps indicates that an amalgamation of local resources was deemed necessary. It must also be

significant that thereafter Muslim authors do not refer to Ajmer until the fourteenth century. A Sanskrit epic

claims that the Chawhan king of Ranthanbor, Hammiradeva (1283-1301), passed

39 TN, II, 72-3 (tr. 839-40); see also I, 491 (tr. 704-6).

40 TFS, 289. The narrower usage is found in TN, I, 284, 400-1 (tr. 200, 468). See generally Yule

and Burnell, Hobson-Jobson, 843-4.

41 Taj, fols. 228a-232b. For Jalor among Iltutmish's conquests, see TN, BL ms., fol. 179b.

42 D. R. Bhandarkar, The Chahamanas of Marwar', EI 11 (1911-12), 74 ff. For Udayasim-ha's

possession of Mandor, see Ray, Dynastic history, 1128-31; and for Iltutmish's campaign, 77V, I, 446 (tr.

611). 43 TFS, 220. FS, 215 (tr. 379-80). TMS, 64, 65. For the kingdom of Jalor, see generally Sharma,

Early Chauhan dynasties, 167-79.

44 TN, I, 468, and II, 8 (tr. 661-2, 728).

~ Map 3a: The war against the Hindu powers in northern India

~ through Ajmer (Ajayameru) in the course of a victorious progress from Abu back to his

capital.45 If reliable, this tradition suggests that the fortress no longer formed part of the Sultanate.

Ranthanbor itself, reputed by Juzjani so impregnable as to have defied the attacks of some seventy Hindu kings down the ages,46 had become tributary in 587/1191. The city was apparently still subordinate

to Iltutmish in 1215, for an inscription at Manglana, on the northern fringes of the Chawhan kingdom,

mentions not only Prthviraja's grandson Valhanadeva but also 'Samasadana' (Shams al-Din) as the

sovereign at Yogini (Delhi). But Ranthanbor had evidently defied the sultan by 623/1226, when his army

captured the fortress after a long siege. Following Iltutmish's death, however, it was invested by the

Hindus, and in c. 635/1237-8 Radiyya sent to its relief the Ghuri malik Qutb al-DIn Hasan b. 'Ali: he seems

to have judged it impossible to hold, since he withdrew the Muslim garrison, dismantled the fortifications

and abandoned the place.47

For several decades the war effort against the Chawhan kingdom did not prosper. At some time

towards the end of Iltutmish's reign, Aytemur al-Baha'i met his death in an expedition from Ajmer into the Bundi region,48 which almost certainly belonged to Prthviraja's descendants. After its recovery by the

Hindus, Ranthanbor was attacked by the Delhi forces on a number of occasions. In 646/1248-9 Ulugh Khan

Balaban led an army towards 'the highlands (kuhpaya) and Ranthanbor' to chastise 'Bahar Deo', described

by Juzjani as 'the greatest of the rais of Hindustan' and identified by modern historians with the Chawhan

king Vagbhata, Valhanadeva's successor. Balaban returned to the attack in 652/1254, during his exile at

Nagawr, and advanced in the direction of 'Ranthanbor, Bundi and Chitor'. On both occasions 'Bahar Deo'

was routed, and the Muslim army obtained a considerable plunder.49 A further campaign was launched

against Ranthanbor early in 657/1259, though the fact that we are told nothing of the outcome is probably

an indication of failure.50

We do not hear of Ranthanbor again in the Muslim sources until Jalal al-Din Khalji's attack in

690/1291, which is described in Amir Khusraw's Miftah al-Futuh. The sultan defeated an army sent against him by the Chawhan king, who then abandoned his capital at Jhayin and fled into the hills. Jalal al-Din

occupied Jhayin, and its idols were smashed to pieces to be taken back to Delhi; detachments were sent on

plundering forays to the

45 Nilkantha Janardan Kirtane, 'The Hammira Mahakavya of Nayachandra Suri', 1A 8 (1879), 64-

5.

46 TN, 1,445 (tr. 610-11).

47 Pandit Rama Kama, 'Manglana stone inscription of Jayatrasimha; (Vikrama-) Samvat 1272', IA

41 (1912), 85-8. TN, I, 445-6, 459-60 (tr. 611, 642). 48 Ibid., II, 8 (tr. 728).

49

Ibid., I, 482-3, and II, 58-9, 65 (tr. 684-5, 818-19, 828); Raverty adopted the reading 'Nahar', but

cf. Ray, Dynastic history, 1095-6.

50 TN, I, 495 (tr. 713).

~ south and east.51

Writing at some remove from these events, Barani gives the impression of a

rather more limited achievement. According to him, Jalal al-Din took Jhayin, demolished its idols and plundered the territory, but spent less than a day inspecting his army's siege operations at Ranthanbor

before raising the investment to avoid further expenditure in Muslim lives. A second campaign against

Jhayin, towards the end of 691/ 1292, accomplished nothing more than the acquisition of booty.52 Apart

from this invasion, the Chawhan kings do not appear to have been seriously troubled by Muslim armies for

the rest of the century, and were free to engage in conflict with their Hindu neighbours, notably the

Paramara kings of Malwa.53 The overthrow of the Chawhans of Ranthanbor was deferred until the reign of

'Ala' al-Din Khalji, and was to herald the Muslim advance on Chitor, Malwa and the lands beyond the

Vindhyas. Not for nothing did Amir Khusraw call Ranthanbor 'the key to the south'.54

Muslim sources barely mention conflict in the thirteenth century with the Guhila kings who

dominated Mewar from their capital at Chitor (now Chittaurgarh), and we are dependent on scattered

references to Muslims in the epigraphy of the region. Of Jaitrasimha (d. just before 1260), the Chirwa inscription of his grandson Samarasimha claims that the Muslims, among others, failed to humble him.

From the same source we learn that during his reign the troops of the suratrana (sultan) attacked

Nagadraha (Nagada). Since Jaitrasimha's resistance to the Muslims elicits high praise in Samar-asimha's

Mount Abu inscription also, it is possible that he was able to avenge this outrage. No further clashes with

the Muslims are reported until the reign of Samarasimha himself (c. 1273-1301), who, in the florid

language of that inscription, 'like unto the primaeval boar ... in a moment lifted the deeply sunk Gurjara

land out of the Turushka sea'.55 Whatever episode is in question here, the statement suggests at least that at

some date prior to 1285 the Guhila monarch had profited from the relative passivity of the Sultanate under

Balaban.

51 MF, 28-35. There is a briefer account in FS, 223-4 (tr. 388-9). Jhayin has been identified by

Satya Prakash Gupta, 'Jhain of the Delhi Sultanate', MIM 3 (1975), 209-15, with Chhain or Chhan, 7 m. S.

of Ranthanbor, at 25° 55' N., 76° 27' E.

52 TFS, 213-14, 220.

53 R. R. Halder, 'Inscription of the time of Hammir of Ranthambor, dated (V.S.) 1345', El 19

(1927-8), 46. On the thirteenth-century Ranthanbor kingdom, see Sharma, Early Chauhan dynasties, 118-

25.

54 KF, 54.

55 Bernhard Geiger, 'Chirwa-Inschrift aus der Zeit des Guhila-Fiirsten Samarasimha rVikrama-

]Sarhvat 1330 [A.D. 1273]', WZKM 21 (1907), 150. F. Kielhorn, 'Mount Abu stone inscription of

Samarasimha [Vikrama-]samvat 1342', IA 16 (1887), 354. See generally Ray, Dynastic history, 1186-90,

1195; but his reference, ibid, 1190, to an appearance in Mewar by Jalal al-Din Mas'ud, brother of Sultan

Nasir al-Din Mahmud, is based on a misreading of CTWR for SNTWR, i.e. Santur (above, p. 73).

~ The Doab and Awadh

In the middle of the thirteenth century the Doab was still largely enemy territory, and Muslim

control in all probability barely extended beyond the walls of the principal towns in the north: Mirat

(Meerut), Kol (close to modern Aligarh) and Baran (now Bulandshahr), all of which had been occupied by

the Muslims since the time of Aybeg. The extent of the problem confronting the government at Delhi is strikingly revealed by the fact that in 647/1249-50 Nasir al-Din Mahmud Shah's forces, on their way to

campaign in the Qinnawj region, were pinned down in warfare with an unidentified Hindu opponent

immediately after crossing the Yamuna.56 The iqta' of Mirat, which was conferred on Balaban's brother

Kishli Khan in 653/1255, is said to have extended 'as far as the foothills of Bandiyaran', and the new muqta'

spent the next few years reducing 'the hills of Bandiyaran as far as Rurki and Mayapur'.57 Early in his reign

Balaban constructed a fortress at Rurki,58 but the place is not mentioned again in the Sultanate period. Nor

is Jhinjhana, which appears as an iqta' during Iltutmish's reign59 and at that time was seemingly the most

northerly Muslim outpost in the region. Nevertheless, the erection of mosques in the 1280s at Manglawr

and Garhmuktesar testifies to the growth of a Muslim presence in the northern Doab.60

The advance of Muslim arms in the southern part of the Doab is better documented. Here certain

towns were clearly under the sultan's rule by the middle of the thirteenth century and are mentioned in iqta'

grants. Maha'tin was joined with Mahir (Mathura), Bhayana and Gwaliyor to form extensive assignments

both for Nusrat al-Din *TaIsI in Iltutmish's day and for Shir Khan in 657/1259. Jalesar and Balaram, the

latter an iqta' already in Radiyya's time, also formed part of Shir Khan's iqta'.61 On the other hand, the very

size of these two extraordinary grants again alerts us to the fact that the Muslims' military resources had to

be stretched over a vast area. Mahir and Maha'un are not heard of again prior to the fourteenth century. It is

possible that the reduction of these tracts had followed upon an obscure victory gained by Iltutmish at

Chandawar in Etawa (scene of the overthrow of the Gahadavala king Jayachandra by the Ghurid forces in

590/1194), to which Juzjani makes one fleeting reference.62 The enemy here

56 TN, I, 483, and II, 60-1 (tr. 686-7, 821). 57 Ibid., II, 47 (tr. 799).

58 FS, 164 (reading ZRKY, but cf. Husain's tr., 291).

59 TN, II, 29 (tr. 759). For Jhinjhana, see DGUP, III. Muzaffarnagar, 263-4.

60 Yazdani, 'Inscriptions of the Turk Sultans', 28-30; ARIE (1975-6), 178 (no. D254); RCEA, XIII

(Cairo, 1944), 21-2 (no. 4832).

61 TN, II, 11, 34, 44, 62, 78 (tr. 733, 767, 794, 824, 849). Balaram is possibly Bilram, 19 m. N. of

Etah, at 27° 49' N., 78° 38' E.: DGUP, XII. Etah, 159-61.

62 TN, II, 17 (tr. 742-3). For Chandawar, 'an ancient village of considerable historical importance',

at 27° 7' N., 78° 23' E., see DGUP, VIII. Agra, 238-9; Irfan Habib, An atlas of the Mughal Empire (Oxford

and Delhi, 1982), 27 and map 8A.

~ may have been Jayachandra's nephew Ajayasimha, who seems to have usurped control over the

Etawa region at least when the rest of the Gahadavala kingdom passed to the late king's son.63 Whatever the

case, local opposition was by no means at an end, and may have revived after Iltutmish's death. In 642/1244

Balaban inflicted severe devastation on the districts of Jarali and Deoli 'and other mawasaf on the northern

borders of Etawa.64 After his accession to the throne, he was obliged to conduct a lengthier campaign in the

Doab, building fortresses at Kampil, Patiyali, Bhojpur and Jalall. Of these, Patiyali, at least, had been in Muslim hands for some time, having been the birthplace of Amir Khusraw in c. 651/1253, and by Balaban's

death had been renamed Mu'minpur (Town of the Faithful').65 Barani mentions that mosques were erected

in these towns, and the date 665 (1267) found on the mosque at Jalall enables us to date the sultan's

operations with some accuracy.66 Such military activity would have made possible the occupation of places

like Sakit (Sekit), where a mosque was raised in 684/1285.67 But Muslim pressure on Etawa was sporadic,

and the assertion of Delhi's authority was a slow process. On its western borders Rapri, lower down the

Yamuna, is not referred to as a Muslim base prior to the reign of 'Ala' al-Din Khalji.68

The result of Balaban's operations in the Doab, says Barani, was to 'open the road to Hindustan'

(i.e. Awadh and the regions to the east).69 One aim had possibly been to safeguard communications with the

isolated strong-hold of Qinnawj, on the right bank of the Ganges. Within a short time of its capture by Aybeg in 595/1199 Qinnawj seems to have been recovered by the Hindus, since Iltutmish is credited with

its conquest and issued coins inscribed 'from the tribute (kharaj) of Qinnawj'.70 Thereafter it served as an

important base of operations against the Hindu princes, its muqta's not only conducting local raids on their

own account71

but also contributing troops, as we shall see, for warfare in 'Malwa' to the south. The four-

month

63 P. Prasad, Sanskrit inscriptions, xviii and 92-4.

64

TN, II, 53-4 (tr. 809). Raverty's identification of the former with 'Jurowli', at 28° 17' N., 78° 17'

E. (809 n.7), is unconvincing. See Hodivala, Studies, I, 230, who suggests either 'Julowlee', 35 m. S. of Fatehgarh, or 'Joolee', 14 m. S. of Sekit; here and at I, 402, he equates Deoli with Thornton's 'Duhlee' in

Etawa, at 27° 2' N., 78° 53' E.: cf. DGUP, X. Mainpuri, 204 ('Dihuli').

65 DGK, 70. Mirza, Life and works, 16-17.

66 TFS, 57-8. For the mosque at Jalall, see Yazdani, 'Inscriptions of the Turk Sultans', 25-6;

RCEA, XII, 122 (no. 4585); for this 'ancient town', 13 m. E. of Aligarh, at 270 52' N., 28° 15' E., see DGUP,

VI. Aligarh, 261-3. Bhojpur lies at 27° 17' N., 79° 41' E., 6 m. S. of Fatehgarh: DGUP, IX. Farrukhabad,

184-5; for Kampil, at 27° 39' N., 79° 20' E., see ibid, 215-16; and for Patiyali, at 27° 42' N., 72° 5' E.,

DGUP, XII. Etah, 201-3.

67 Yazdani, 'Inscriptions of the Turk Sultans', 31-2; RCEA, XIII, 47-8 (no. 4870). 68 TFS, 328, 333; and see Hodivala, Studies, I, 281. G. H. Yazdani, 'Inscriptions of the Khalji

Sultans of Delhi and their contemporaries in Bengal', EIM(1917-18), 30.

69 TFS, 58.

70 CCIM, I, part 2, 21 (no. 39). CMSD, 71-2 (no. 52). Taj, in ED, II, 241. 77V, I, 452 (tr. 627).

71 Ibid., I, 470 (tr. 665).

~ campaign of 647/1249-50 by the army of Nasir al-Din Mahmud Shah appears to have been directed at the Qinnawj region, for an inscription from Bilram records the martyrdom of a Muslim

commander at this time at a village near Qinnawj.72 It may similarly have taken some considerable time to

wrest the region between Qinnawj and the distant outpost of Kara from the infidel. *Barihun (Barhamun?),

which is known to have lain in the vicinity of Qinnawj and which was twice granted as iqta' by Iltutmish, is

not mentioned again.73 The walls of Tilsanda, which JuzjanI likens to Alexan-der's Gates, succumbed only

in 645/1248 to an attack by Mahmud Shah's forces.74 And there were in all likelihood reverses, which

claimed the lives of the Muslims commemorated in various epitaphs from Bilram dated 658/ 1260,

683/1284 and 703/1303, the last of them in the fortress of Chandawar.75

East of the Ganges, the important stronghold of Bada'un, in Muslim hands since 594/1198, faced

towards Katehr (corresponding roughly to present-day Rohilkhand), which surfaces repeatedly in our

sources as the territory of the most refractory of infidel peoples. The reduction of this tract by Iltutmish, to which JuzjanI alludes, may have occurred in 1227.76 But it was a shortlived affair, since Sanjar-i

Qabaqulaq, as muqta' of Bada'un under Mas'ud Shah, is said to have overthrown 'the mawasat of Katehr

and Bada'un', conducted numerous expeditions and founded mosques.77 Around the middle of the thirteenth

century we hear of other Muslim-held centres in these parts, like Sambhal and Kasrak, which first appear as

iqta's in Mahmud Shah's reign, and Amroha, which became an iqta' only under Balaban.78 In 652/1254,

during a campaign that had taken him through Bardar (Hardwar) and Bijnor as far as the banks of the

Rahab (Ramganga), Mahmud Shah retaliated for the loss of one of his lieutenants by inflicting, in Juzjani's

words, 'a reverse on the infidels of Katehr that [the people of] that territory will remember for the rest of

their lives'.79 But memory, it

72 ARIE (1966-7), 82 (no. D249). For the expedition, see TN, I, 483, and II, 60-1 (tr. 686-7, 821). 73 Ibid., II, 20, 36 (tr. 746, 779). This place is mentioned by tenth-century geographers: al-

Muqaddasi, Descriptio imperil Moslemici, ed. M. J. De Goeje (Leiden, 1877), 478 (variant reading:

BRHYN); Hudud al-'Alam, facs. edn V. V. Bartol'd (Leningrad, 1930), fol. 15b (tr. Minorsky, 90), placing it

near Qinnawj.

74 TN, I, 480, 481 (tr. 679-80, 681). See Raverty, 679-80 n. 6, and Hodivala, Studies, I, 222: the

village of Tilsanda near Cawnpore (Kanpur), mentioned by the latter, seems most likely.

75 ARIE (1966-7), 83 (nos. D252, D253, D255). 76 When Iltutmish is known to have been at Bada'un: P. Prasad, Sanskrit inscriptions, 80-9. Katehr

is omitted in the printed edition of TN, but cf. BL ms., fol. 180a (corrupted to 'LYHR). The varying

significance of the term 'Katehr' is discussed in Hodivala, Studies, I, 259.

77 TN, II, 26 (tr. 755).

78 Sambhal: ibid, I, 482 (tr. 684). Amroha: TFS, 36-7, and Habibullah, Foundation, 156. Kasrak:

TN, II, 42 (KRK; but cf. BL ms., fol. 211a, KNRK, and Raverty's tr., 791); the place is identified by P. Prasad,

Sanskrit inscriptions, 81-2, with a village in Tilhar tahsil in Shahjahanpur district.

79 TN, I, 487-8 (tr. 696-8).

~ Map 3b: The war against the Hindu powers in northern India

~ seems, was brief. The disorders of c. 665/1266-7, in which Bada'un, Sambhal, Amroha and

Ganuri were subject to further depredations by the infidel, brought down upon Katehr the wrath of Balaban,

who razed the town of that name and despatched five thousand archers to ravage the rest of the territory. As

a result, claims Barani, there was no further trouble from Katehr until the end of the reign of Jalal al-Din

Khalji.80 During his time Ganuri was joined with Chawpala (present-day Moradabad) on the Rahab to form

a sizeable iqta', indicating perhaps that resources here were thinly spread.81 Yet Jalal al-Din mounted

plundering operations in the Kabar district, and under 'Ala' al-Din the government's grip was sufficiently

secure for it to be incorporated into the khalisa.82

South of Bada'un lay the extensive territory of Awadh. The Muslims had at some unspecified

point occupied the city of this name, the ancient Ayodhya, which in the 1190s was the iqta' of Husam al-

Din Oghulbeg. Early in the thirteenth century we hear of other towns in the area which served as iqta'

headquarters, like Kasmandi and Mandiyana (Mandiaon), both in the neighbourhood of the later city of

Lakhnaw (Lucknow).83 The muqta's of Awadh played a prominent role in the interminable war against the

infidel. It was while he was based in Awadh that not long after 623/1226 Iltutmish's eldest son, Nasir al-

Din Mahmud, crushed a ruler 'Hardu Dal', called by Juzjani 'Bartu' (possibly a subordinate of the

Gahadavalas), who had been a thorn in the flesh of the Muslim settlers in the region. But the prince's

victory, commemorated in verses of the poet Siraji,84 did not effect the pacification of Awadh, and in the

following decades Taj al-DIn Sanjar *Kirit Khan and Taj al-Din *Teniz Khan are each in turn credited with

uprooting the mawasat of the region.85 Thereafter, although we learn at intervals the names of Awadh's muqta's, including for a short time the future sultan 'Ala' al-Din, there is a gap in our knowledge of

developments here until the Tughluqid period.

Banaras had been taken from the Gahadavalas in 590/1194, and Muhammad b. Bakhtiyar had

profited from their troubles to establish a base in the districts of *Sekhit and Bhoili, in the Chunar region on

the opposite bank of the Ganges, from where he mounted operations against

80 TFS, 59. Ganuri is identified by Hodivala, Studies, I, 259-60, with Thornton's 'Genori' or

'Genouri', in Bulandshahr, at 28° 20' N., 78° 4' E.

81 TFS, 204 (KANWD, to be corrected from BL ms., fol. 110b). For Chawpala, see MF, 13;

Hodivala, Studies, I, 366; DGUP, XVI. Moradabad, 232.

82 MF, 21-2. TFS, 323-4. For Kabar, later renamed Shergarh and 'an ancient town', at 28° 39' N.,

79° 22' E., see DGUP, XIII. Bareilly, 268-9.

83 TN, II, 29, 72 (tr. 759, 838). For Kasmandl, see Hodivala, Studies, I, 233, and DGUP, XXXVII.

Lucknow, 195-6. For Mandiaon, at 26° 56' N., 80° 58' E., see ibid, 245-7; Habib, Atlas, map 8A.

84

Siddiqi, 'Historical information', 55-6. TN, I, 453 (tr. 628-9). For a different identification, see

Habibullah, Foundation, 111 n.99. Hodivala, Studies, I, 218, was surely wrong to link 'Bartu' with a raja in the Bhitargarh district of eastern Bengal.

85 TN, II, 27, 29 (tr. 757, 760).

~ Maner and Bihar.86 But the Muslim foothold here may have been transient. The Gahadavalas

continued to hold out in the more inaccessible parts of the Banaras and Chunar regions;87 and although

JuzjanI mentions Banaras among the strongpoints taken by Iltutmish from defeated Muslim rebels (p. 29

above), it is not mentioned as an iqta' again and disappears from sight. The first inscription attesting

Muslim rule belongs to the reign of Qutb al-DIn Mubarak Shah Khalji (716-20/1316-20).88

The road to Bengal In the early thirteenth century the furthest Muslim base beyond Awadh in the direction of the

infidel territories of Nepal and Tirhut was the old town of Bahraich, which housed the shrine of Salar

Mas'ud, a Muslim warrior of uncertain background, and had become a major focus of pilgrimage.89

Bahraich was held by the Muslims in Iltutmish's reign; and here, as JuzjanI proudly announces, his

sovereign Nasir al-Din Mahmud Shah waged holy war against unspecified Hindus prior to his accession.90

It may have been the same regions that in 690/1291 were ravaged by Jalal al-DIn Khalji following his

victory over the rebel Malik Chhajju, for Amir Khusraw says that in cutting down the jungles he was

seeking to open a road to Lakhnawti: mawas after mawas yielded tribute to the sultan, but regrettably none

of the places Khusraw mentions can be identified.91 Barani asserts that the rai of Gorakhpur had paid tribute

to the administrative district (shiqq) of Awadh prior to the upheavals of Muhammad b. Tughluq's reign;92

but it is doubtful if the relationship went back into the thirteenth century.

During the early Sultanate period Muslim armies were raiding both Tirhut (north Bihar) and the

neighbouring region of Nepal known in Juzjanl's time as 'Bhatlgun'. Tirhut had paid tribute to Ghiyath al-

Din Twad, which is doubtless why JuzjanI lists it among Iltutmish's conquests,93

86 Ibid., I, 423 (readings corrupt; Raverty's tr., 549-50, has 'Bhagwat' for the first, but cf. BL ms.,

fol. 170a, SKHYT): see Hodivala, Studies, I, 206, on the identification of these places, of which the second,

Thornton's 'Bhoelee', lies at 25° 6' N., 83° 3' E.; also DGUP, XXVII. Mirzapur, 278-84.

87 P. Prasad, Sanskrit inscriptions, 56-70 (nos. 11:5-6). See more generally Niyogi, History of the

Gahadavala dynasty, 113-19.

88 ARIE(\91\-2), 91 (no. D170).

89 I. H. Siddiqui, 'A note on the Dargah of Salar Mas'ud in Bahraich in the light of the standard

historical sources', in Christian W. Troll (ed.), Muslim shrines in India (Delhi, 1989), 44-7.

90 Iltutmish: 'Awfi, Lubab al-Albab, I, 115. Mahmud Shah: TN, I, 470, 478 (tr. 665, 676).

91 MF, 22-3 (reading KSHWN for KYTHWN, i.e. *Kaithun). FS, 224 (tr. 390-1), with 'BRY and

KYTHWN. TMS, 64, includes a faint echo of this campaign ('NHRY KYTHWR) in an obviously corrupt

sentence.

92 TFS, 587-8. SFS, 33, likewise claims that 'Karosa' and Gorakhpur were dependencies of Awadh.

93 TN, I, 437 (tr. 587-8); omitted in the list of Iltutmish's conquests at I, 452, but see BL ms., fol.

180a.

~ and in Radiyya's time Toghril Toghan Khan, as muqta' of Bihar and Lakhnawti, conducted a

lucrative raid on Tirhut.94 But Tirhut was also vulnerable to attacks from the opposite direction. Radiyya's

muqta' of Awadh, Temur Khan, penetrated 'as far as the limits of Tirhut' and exacted tribute from

Bhatigun;95

and at the beginning of 654/1256 Balaban, on the heels of his enemy Qutlugh Khan, led an

army as far as 'Bhatigun and the limits of Tirhut', returning with a vast plunder.96 In 702-3/1302-3, the muqta' of Kara is found commanding the troops of 'the east, Bengal and Tirhut',97 which is a mystery, given

that Bengal did not then form part of the sultan's dominions. Otherwise we hear no more of Muslim attacks

on Tirhut until Ghiyath al-Din Tughluq's expedition of 724/1324.

The subjugation of Bihar, of which Juzjani gives no details apart from an account of the capture of

the city itself (identical, in all probability, with Uddandapuri),98 was clearly a piecemeal process. Gaya

(Vajrasana), which was under Muslim rule in 1219,99 was no longer in Muslim hands when the Buddhist

monk Dharmasvamin passed through Bihar in 1234; although he bears eloquent testimony to continued

Muslim military activity in both Bihar and Tirhut.100 Following Balaban's recovery of the eastern provinces

soon after c. 665/1267, Gaya may still have been unsubdued, since a local ruler boasts in an inscription of

1268 that he has preserved his territory from the sultan.101

Much of the credit for the implantation of Islam in west Bengal belongs to the Khalaj rulers, who

are praised for building mosques, colleges (maddris) and hospices (khanaqahat), and for munificence

towards Muslim scholars and sayyids.102 A paucity of sources, however, renders the progress of Muslim

settlement here even more obscure than it is elsewhere. After Juzjani's time, inscriptions show that one of

the Muslim strongpoints he

94 TN, II, 14 (tr. 737).

95 Ibid., II, 17-18 (tr. 743). Here (n.3) Raverty, who had earlier identified this region correctly

(ibid, 639 n.8; cf. also 567 n.l), defined it as 'the tract lying on the left bank of the Son, east of Banaras'. Cf.

also Habibullah, Foundation, 150, who, following Cunningham, defines it as the Tons valley ('Bhath-gora'). But this is to ignore the link with Tirhut which Juzjani makes on both occasions. The name is

perhaps connected with the town of Bhatgaon, on which see IG, VIII, 89.

96 TN, II, 71-2 (Raverty's tr., 838, reads 'Badikot', but the place is evidently identical with that

mentioned in Temur Khan's biography).

97 TFS1 , Digby Coll. ms, fol. 113a; TFS, 300.

98 TN, I, 423 (tr. 551-2); a brief reference to the conquest of the territory of Bihar at I, 425 (tr.

556). For Uddandapuri, see Shri Hasan Nishat Ansari, 'Historical geography of Bihar on the eve of the

early Turkish invasion', JBRS 49 (1963), 257 n.3.

99 ARIE (1962-3), 24, 80 (no. B261).

100 G. Roerich, The biography of Dharmasvamin (Chag lo-tsa-ba Chos-rje-dpal) (Patna, 1959),

61-4. For the date, see the introduction by A. S. Altekar, v-vi; and for Muslim pressure on Bihar, ibid., xix.

101 Hasan Nishat Ansari, 'Gaya epigraph of V.s. 1325 noticing Balban as Biruban', JBRS 53

(1967), 170-81: the editor is sceptical and suggests that this security was purchased with tribute.

102 TN, I, 427, 436 (tr. 559-60, 583).

~ mentions - Deokot, which was the capital until 'Iwad's accession - was still a centre of Muslim power at the end of the thirteenth century, under the independent Ghiyathid sultans; but others - like

Narangui, Ganguri, or Basankot (the last-named founded by 'Iwad) - do not surface again.103 Juzjani

distinguishes the region of Ral (Rarh), with its centre at Lakhnor, west of the Ganges, from Barind

(Varendra), the tract to the east of the river, which included Deokot.104

After its capture by the Hindus in

642/ 1244 (below), we simply do not know when Lakhnor was recovered.

'Iwad extorted tribute from the Sena kings in eastern Bengal (Bang), and after the conquest of

Lakhnawti by Iltutmish this pressure was maintained by his representatives. The second of these, Sayf al-

Din Aybeg, owed his sobriquet Yaghantut to the great number of elephants he obtained by way of plunder

from Bang and despatched to the sultan.105 Toghril Khan Ikhtiyar al-Din Yuzbeg is found in 653/1255 striking coins from 'the revenue (kharaj) of the territory ('ard) of *Badar and Nudiya';106 the mention of the

latter city demonstrates that the Senas had at some point reoccupied their old capital. They survived into the

second half of the century, when they were apparently supplanted by a rival power which had arisen in

Tipperah. This region, which Barani calls 'Jajnagar', may have been the object of a lucrative expedition by

Balaban's lieutenant Toghril, as governor of Lakhnawtl, immediately prior to his rebellion against the

sultan and his assumption of the imperial title.107 One of the two known kings of the Tipperah dynasty was

the 'Danuj Rai of Sunarga'un (Sonargaon) who sought his revenge by cooperating with Balaban and

undertaking to obstruct the rebel's flight.108

Bang was only one of the territories that offered rich pickings to the Muslim rulers of Lakhnawtl.

Nobody sought to emulate Muhammad b. Bakhtiyar, who had returned a broken man from a disastrous

campaign through the upper Brahmaputra region, possibly into the region of Assam the Muslims called Kamrtup or Kamrud;109 but his successors at Lakhnawtl

103 Basankot: ibid, I, 436 (tr. 581-2). Narangui: I, 432, 434 (tr. 572, 574, 577). Ganguri: I, 433 (tr.

575). Deokot: I, 431-4 (tr. 572, 574, 576, 578); Abdul Karim (ed.), Corpus of the Arabic and Persian

inscriptions of Bengal (Dacca, 1992), 48-53. On the possible location of some of these places, see

Monmohan Chakravarti, 'Notes on Gaur and other old places in Bengal', JASBns5(1909), 199-235.

104 TN, I, 436-7, and II, 13 (tr. 584-5, 737).

105 Ibid., II, 9-10 (tr. 732). For Tu. yaghantut, 'seize elephants', see Clauson, Etymological

dictionary, 904, in conjunction with Rasonyi, 'Noms de personnes', 241-2. 106 CCIM, II, 146 (no. 6). CMSD, 55 (no. 225D).

107 TFS 82; 83. For Barani's usage (Jahajnagar, 'city of boats or ships'), see Chakravarti, 'Notes on

Gaur', 217.

108 TFS, 87 TMS, 42-3. He has been identified with Ariraja Danujamadhava Dasarathadeva, who

issued a copper-plate grant around this time: Habibullah, Foundation, 183 n.32; Ray, Dynastic history, 383

n.l; and for a survey of the later Senas, ibid., 379ff.

109 TN, I, 427-32 (tr. 560-72). On this expedition, see the evidence reviewed in Digby, War-horse

and elephant, 45-6; cf. the dubious hypothesis in Z. V. Togan, 'About the campaign of the Indian Khalach-Turks against the Keraits of Mongolia in Northern Tibet in the

~ conducted operations over a vast area. Iwad levied tribute not merely on Kamrup but also on the

Eastern Ganga kingdom of Orissa, usually designated in the Muslim sources as Jajnagar.110 On balance the

Delhi Sultan's governors were less successful. Toghan Khan was in 641/1244 worsted in continual warfare

with the king of Jajnagar, with the result that a Hindu army took Lakhnor, where Toghan Khan's lieutenant

fell in the fighting, and menaced Lakhnawti itself: it was saved only by the timely appearance of troops sent

by the Delhi Sultan (above, p. 91).111 The reverse at the hands of Jajnagar appears to be celebrated in an

inscription of King Narasimha II of Orissa, referring to his father's victory over the 'Yavanas of Radha

(Rarh) and Varendra'.112 Toghan Khan's humiliation was avenged some ten years later by Yuzbeg, who

sacked the Jajnagar king's capital at an unidentified place named '*Umardan'.113 But Yiizbeg himself met his death in c. 655/1257 while engaged in an ambitious and heedless invasion of Kamrup.114 From a fath-

ndma in Amir Khusraw's Rasa'il al-I'jaz, it appears that in 680/1282 Balaban himself led a campaign into

Orissa which reduced King 'Mal Deo' to submission.115

The era of the independent Ghiyathid sultans and of their successors of the line of Firaz-i Aytegin

was marked by significant advances at the expense of the independent Hindu powers within Bengal proper.

From a coin issued by Balaban's grandson Kayka'us and struck 'from the kharaj of Bang', it is clear that

part of the eastern delta was now once again tributary to the Muslims.116 Under Shams al-Din Firaz Shah,

Sunarga'un appears as a Muslim mint-town. Satga'un (Satgaon) too appears to have been annexed in his

reign, and a mosque was built at Tribeni, on the Hooghly, in 698/ 1298-9. Epigraphical evidence reveals that Sirihat (Sylhet), lying immediately east of the Brahmaputra, was reduced in 703/1303-4.117 All these

places would pass briefly under the rule of the Delhi Sultan as a result of Ghiyath al-DIn Tughluq's

expedition of 724/1324 (below, pp. 200-1).

years 1205-1206', JPHS 12 (1964), 187-94, and in Proceedings of the 36th International Congress

of Orientalists (Delhi, 1968), 174-8.

110 TN, I, 437 (tr. 587-8).

111 Ibid., II, 15 (tr. 739-40).

112 N. Vasu, 'Copper-plate inscription of Nrsimha-deva II of Orissa, dated 1217 Caka', JASB 65,

part 1 (1896), 232-4, 267 (verse 84).

113 TN, II, 31 (tr. 762-3). Habibullah, Foundation, 144, was sceptical that '*Umardan' was actually

deep within Orissa.

114 TN, II, 32-3 (tr. 764-6).

115 RI, V, 5-13: see 8-11 for 'Mal Deo'; 13 for the date; 'WDH must be an error for 'WRSH.

Khusraw's heading for the fath-nama refers only to Lakhnawti, which misled Mirza, Life and works, 219.

116 H. E. Stapleton, 'Contributions to the history and ethnology of north-eastern India - IV. Bengal

chronology during the period of independent Muslim rule. Part I, 686-735 A.H. (1286-1334 A.D.)', JASB

ns 18 (1922), 410.

117 Abdul Karim, Corpus, 53-6. Stapleton, 411-12. For the Sylhet inscription, see Ahmad Hasan

Dani, Muslim inscriptions of Bengal (Dacca, 1957), 7 (no. 9). See generally Majumdar, History of medieval

Bengal, 17-19.

~ Bundelkhand and Malwa

The town of Bhayana originated with the settlement of Sultankot, founded by Baha' al-Din Toghril

as part of his strategy to take Gwaliyor.118 Gwaliyor, which had submitted to Aybeg in 597/1200-1, was

retaken by the Hindus, doubtless after Aybeg's death, but in 630/1233 the city fell to Iltutmish after an eleven-month investment; the last Pratihara (Parihar) king, 'Mangal Deo', fled to Narwar.119 Nusrat al-Din

*TaIsI, whom Iltutmish in 631/1234 appointed as muqta' of Bhayana and Sultankot and prefect (shihna) of

Gwaliyor, was in addition entrusted with overall command of the troops of Qinnawj, Mahir and Maha'un,

perhaps an indication of the importance attached by the sultan to this particular front.120 From these bases

Muslim commanders waged war on the Chan-della kingdom of Jejakabhukti (modern Bundelkhand). Its

capital, Kalinjar, had fallen to Qutb al-Din Aybeg in 599/1203, following the death of King Paramardldeva,

and had briefly been an iqta'121 But Trailokyavarman, Paramardideva's son and heir, had evidently

recovered it by Iltutmish's last years.122 *TaIsI defeated Trailokyavarman, capturing the rai's ceremonial

parasol and standards during the pursuit.123 Yet Muslim military activity had little impact on the Chandella

kingdom. Although Kalinjar itself was not re-established as the capital following its recovery, and the kings

normally resided at Ajayagarh, some twenty miles to the south-west, this seems to have been the sole concession made to the proximity of Muslim power by the Chandella ruler, who continued to style himself

'king of Kalinjar'. In an inscription of his son Viravarman, Trailokyavarman is credited with 'lifting up the

earth immersed in the ocean formed by the streams of the Turushkas'.124 VIravarman himself issued a

copper-plate grant in 1280 to a nobleman who had vanquished the 'Turushkas' among others.125

Nor were

these monarchs deflected from the duty of warring against their Hindu neighbours, notably the Kalachuri

kings of Chedi and the Jajapellas of Narwar.126

118 TN, I, 421 (tr. 545); cf. also II, 34 (tr. 767).

119 Ibid., I, 448 (tr. 619). The form adopted by Raverty is 'Mangal-deV, though BL ms. reads

MYKL DYW; another variant reading, MLKDYW, suggests something like 'Melugideva' (cf. Hodivala, Studies,

I, 216). For the last Pratihara kings, see Ray, Dynastic history, 829 n.l, 834 n.l.

120 TN, II, 10-11 (tr. 732-3). 121 Taj, fols. 182a-185a.

122 TN, II, 10-11, 12 (tr. 733, 734-5). K. N. Dikshit, 'Garra plates of the Chandella

Trailokyavarman: [Vikrama-]samvat 1261', El 16 (1921-2), 272-7. R. K. Dikshit, Candellas, 157-8,

following Ray, Dynastic history, 725, suggests that Trailokyavarman recovered Kalinjar as early as 1205.

123 TN, II, 10-11, 12, 62-3(tr. 733, 734-5, 834-5).

124 P. Prasad, Sanskrit inscriptions, 100-5 (verse 7). 125 Sir Alexander Cunningham, 'Report of a tour in Bundelkhand and Rewa in 1883-84', ASIR 21

(1885), 75. The grant is no longer extant: Ray, Dynastic history, 732, where the date cited, Vikrama samvat

1237, is an error for 1337.

126 For the attack on Narwar, see D. C. Sircar, 'Inscriptions of the time of Yajvapala Gopala',

~ During the return march from his invasion of the Chandella kingdom in 632/1235, *Taisi had

been ambushed by a ruler whom Juzjani calls 'Chahar-i Ajari'. This was the earliest recorded Muslim clash

with Chahadadeva, the second of the Jajapella (Yajvapala) kings, who around this time wrested the

stronghold of Narwar from the Pratiharas and made it their residence. Juzjani, who was then qadi of Gwaliyor, heard an account of the engagement from the veteran amir's own mouth, and it is clear that

*Taisi had extricated his army with considerable difficulty.127 He died soon after Radiyya's accession, and

in 635/1238 her troops evacuated the Muslim population of Gwaliyor.128 Juzjani retained the office of qadi,

and this was confirmed in 643/1245, but on each occasion he simultaneously received an important post in

Delhi, suggesting that the Gwaliyor appointment was simply one in partibus infidelium.129 That the great

fortress was now in enemy territory emerges from the campaign launched at some point in Radiyya's reign

by Temur Khan Qiran from Awadh towards 'Gwaliyor and Malwa' in which he is said to have done signal

service.130

The enemy was undoubtedly the Jajapella king. Juzjani, describing Ulugh Khan Balaban's

campaign against him in 649-50/1251-2, says that it headed towards 'Gwaliyor, Chanderi, Nurwul (Narwar)

and Malwa' and speaks of him as 'the greatest of all the rais of that country'. Balaban succeeded in taking Narwar and putting it to the sack.131 Gwaliyor seems to have been recovered at this juncture, for in

657/1259 the governorship of the fortress, together with a large iqta' comprising Kol, Bhayana, Balaram,

Mahir, Maha'un and other territories, was conferred on Balaban's cousin Shir Khan.132 Once again, the

amalgamation of widely dispersed resources may indicate both the fragility of Muslim rule and the

magnitude of the Hindu threat. Gwaliyor, which Juzjani could call 'one of the celebrated strongholds of

Islam', was apparently in Hindu hands once more by the last years of Balaban's reign, for among those

overcome by the noble who received the above-mentioned grant from the Chandella monarch Vira-varman

in 1280 was Hariraja of Gopagiri, i.e. Gwaliyor.133 Unless Hariraja

El 31 (1955-6), 326-7; and for relations with the Jajapellas generally, see R. K. Dikshit, Candellas,

168-70. 127 TN, II, 11-12, 62-3 (tr. 733-4, 824-5). For the correct identification of 'Chahar', see M. B.

Garde, 'A note on the Yajvapalas or Jajapellas of Narwar', IA 47 (1918), 241-4, refuting the suggestion that

he belonged to a branch of the Chawhans; though the older view is reiterated by Ray, Dynastic history,

1103-4. The surname 'Ajari' is discussed by Hodivala, Studies, I, 224-5, who connects it with the locality

Arjar, some 40 m. S.E. of Narwar.

128 TN, II, 25-6 (tr. 754-5). Mongol envoys incarcerated at Gwaliyor were transferred to Qinnawj:

ibid., II, 214 (tr. 1285). For Taisi's death, see ibid, I, 458, and II, 13 (tr. 639-40, 735-6).

129 Ibid., I, 460, 470 (tr. 644, 667).

130 Ibid., II, 17 (tr. 743).

131 Ibid., I, 485-6, and II, 62, 63 (tr. 690-2, 824-5).

132 Ibid., II, 44 (tr. 794); and for the date, cf. I, 495 (tr. 712-13).

133 Cunningham, 'Report of a tour in Bundelkhand and Rewa', 74-6. For the phrase from Juzjani,

see TN, II, 44 (tr. 794).

~ was a client ruler under the sultan's overlordship, which is unlikely, then the grant shows that the fortress had once more slipped out of Muslim control. This possibility is borne out by the fact that Jalal al-

Din Khaljl is found conducting a plundering expedition to Gwaliyor shortly before his murder in

695/1296.134

As for Chahadadeva, little reliance can be placed on Juzjani's assertion that he had been totally

overthrown and rooted out of his kingdom. The coins and inscriptions of the Jajapella dynasty dictate a

more sober assessment, demonstrating as they do that Chahadadeva and his successors maintained their

hold on Narwar into the fourteenth century. The ultimate fate of the Jajapellas is obscure, and it is usually

assumed that they fell victim to an unrecorded invasion by the forces of 'Ala' al-Din Khaljl.135 It would

have been some such campaign that resulted in the Muslim occupation of Chanderi, which is known to

have occurred prior to 711/1312.136

To the north-east of the Jajapella and Chandella kingdoms lay the Muslim outposts of Kara and

Manikpur, which are frequently linked in the sources. Kara acknowledged Muslim authority as early as

Aybeg's reign, and was the seat of an amir in the time of Iltutmish.137 When muqta' of Kara during

Radiyya's reign, Temur Khan Qiran is said to have conducted numerous forays against the infidel, but no

details are given.138 In 645/1248 we find the army of Nasir al-Din Mahmud Shah based at Kara, while

Balaban led a detachment against a Hindu potentate called by Juzjani 'Dalakimalaki' (?). The dynastic

affiliations of this prince, said to have occupied the regions along the Yamuna between Kara and Kalinjar,

are uncertain; but he was not, apparently, a Chandella, for we are told that 'the rais of the marches (atraf) of

Kalinjar and Malwa' were unable to subdue him. His stronghold was looted by the Muslims, and his family

and dependants were captured.139 This was the victory which Juzjani tells us he commemorated in his lost

Nasiri-Nama; but for all the chronicler's bombast it is difficult to withstand the impression that the campaign was on a relatively trifling scale.140 We do not hear of operations from Kara again

134 TFS, 223, 228. TMS, 67; for Sirhindi's claim that Gwaliyor was in Muslim hands again by the

first year of 'Ala' al-Din's reign, see ibid, 72 (the text reads KALPWR).

135 Sircar, 'Inscriptions of the time of Yajvapala Gopala', 323-36; also his 'Yajvapala Gopala', IHQ

32 (1956), 399-405.

136 Z. A. Desai, The Chanderi inscription of' 'Alau'd-din Khalji', EIAPS (1968), 4-10. Cf. also TFS,

323.

137 ARIE (1969-70), 11, 98 (no. D214). TN, I, 452 (reading KWH for KRH; but cf. Raverty's tr., 627).

On Kara, see generally Laiq Ahmad, 'Kara'.

138

TN, II, 17 (tr. 743).

139 Ibid., I, 481-2, and II, 57-8 (tr. 681-3, 816-18). Ray, Dynastic history, 729-30, and Hodivala,

Studies, I, 222-3, thought 'Dalakimalaki' might be the Chandella king Trailo-kyavarman; but Juzjani's

account suggests a less powerful chief.

140 Laiq Ahmad, 'Kara', 84, assumes that Kishli Khan was appointed as muqta' of Kara in 653/1255

(recte 651/1253) and that for a few years prior to this it was in Hindu hands.

~ until 'Ala' al-Din Khalji became muqta' of the region in 689/1290 and 'trampled underfoot

numerous mawasaf .141

According to Juzjani, Aybeg had conquered the territory as far as the frontiers of Ujjain.142 The

Sultanate's first war against the Paramara dynasty of Malwa was deferred until the end of the reign of

Iltutmish, who in 632/ 1235 invaded the kingdom and successively plundered the cities of Bhilsan (Bhilsa)

and Ujjain, destroying the temple at the former place and removing the idol of Mahakala to Delhi.143 Like

so many other exploits of the period, however, this campaign had no permanent results; and we have no

record of any subsequent Muslim attack on the Paramaras before the Khalji era, when in c. 692/1293 the

future sultan 'Ala' al-DIn, as muqta' of Kara, plundered Bhilsan and carried off its great bronze idol.144 Juzjani speaks of campaigns towards Malwa on a number of occasions, but he is not employing the term in

its narrow sense, to refer to the Paramara kingdom, which in his day was beyond the reach of the sultan's

lieutenants. In the same vein Amir Khusraw describes the troops of Jalal al-Din Khalji as advancing as far

as the borders of Malwa in 690/1291, when they crossed the Chambal and the Kunwari (Kunar).145 For

these writers, 'Malwa' appears to function as a general label for the entire region lying south and south-west

of modern Bundelkhand.

The prospect of dominion

The idea of paramountcy over the entire subcontinent had a long pedi-gree,146 and had possibly

communicated itself to Delhi's sovereigns. A Sanskrit inscription of Balaban's reign might have appeared to give them every encouragement, since it depicts the sultan's authority as radiating over 'the Dravida country

and Rameshvaram'.147 Court poets, too, flattered Muslim rulers regarding their putative conquests. In the

Qiran al-Sa'dayn, Mu'izz al-DIn Kayqubad is made to boast: 'Sometimes I give my troops gold from

Gujarat; at others I write drafts [for them] on Deogir ... I make Malwa the repository of my riches; Jajnagar

I cause to meet the obligations of my treasury ... '148 At the time, however, this was mere fantasy. Prior to

141 KF, 8; see also FS, 227-8 (tr. 393-4), for his severity towards recalcitrant Hindus while at Kara.

142 TN, I, 417 (tr. 516-17); and see above, p. 19, n.62.

143 TN, I, 449 (with the year 631), 452 (cf. BL ms, fol. 180a; tr. 621-3, 628). Ray, Dynastic history,

907, places this invasion in the reign of Devapala (c. 1218-36). 144 TFS, 220.

145 MF, 34.

146 Andre Wink, Land and sovereignty in India (Cambridge, 1986), sect. I, esp. 15-21.

147 Palam Baoli inscription, V.s. 1333/1276, in P. Prasad, Sanskrit inscriptions, 5-6, 12 (no. 1:4).

See the comments of Peter Hardy, 'The authority of Muslim kings in mediaeval South Asia', in Marc

Gaborieau (ed.), Islam et societe en Asie du Sud, Collection Purusarthe 9 (Paris, 1986), 39.

148 QS, 63, gah ba-hasham zar diham az Gujarat * gah ba-Diwglr nawisam barat * ... Malwa-ra

waqj'-i dafa'in kunam * Jaj [printed text has jam in error] nagar wajh-i khazain kunam (tr. in ED, III, 526).

~ 'Ala' al-Din's audacious raid of 695/1296, the people of Deogir had never even seen a Muslim

army from the north.149 For the Muslims of the thirteenth-century Sultanate, most of central and peninsular

India was terra incognita. In this direction the Hindu territories seemed to stretch away indefinitely. The

bounds of Malwa, wrote Amir Khusraw later - again fully availing himself of poetic licence - exceeded the

ability of skilled surveyors (muhandisan) to measure them.150 It is a striking testimony to the vigour of 'Ala'

al-Din's regime that, as we shall see in chapter 10, events had already overtaken this observation some

years before Khusraw died. 149 TFS, 222-3. For Hindu poetry and epigraphy of the Yadava kingdom which seems to contradict

Barani, see P. M. Joshi and A. Mahdi Husain, 'Khaljis and Tughluqs in the Deccan', in H. K. Sherwani and

P. M. Joshi (eds.), History of medieval Deccan (1295-1724) (Hyderabad, A. P., 1973, 2 vols.), I, 34-5, who

suggest, however, that these refer to some clash with Muslims in the coastal region.

150 KF, 56.

~ PART II

The zenith of the Sultanate ~ CHAPTER 8

Sultans, saints and sources

Sources for the period down to 752/1351

For the whole period from 'Ala' al-Din's reign (695-715/1296-1316) through to the early 1350s, we

continue to be dependent largely on three authors writing within India, namely Barani, 'Isami and Sirhindi,

together with - for Muhammad b. Tughluq's reign (724-52/1324-51) - the memoirs of the Moroccan visitor

Ibn Battuta. Authors writing in the Mughal era -notably Nizam al-Din Ahmad Harawi (d. 1003/1594), 'Abd

al-Qadir Ba da'uni (late sixteenth century), and the seventeenth-century compilators Ulughkhani (Hajji al-Dabir) and Muhammad Qasim Hindu Shah Astar-abadi (Firishta) - generally rely upon Barani or Sirhindl

and have no value as primary sources; but they occasionally preserve for us details gleaned from earlier

works that are no longer extant. It should be noticed that there exists an earlier version of Barani's Ta'rikh-i

Firuz-Shahi (utilized by the fifteenth-century writer Bihamadkhani), of which there have survived three

manuscripts1 and which from 'Ala' al-Din onwards begins to diverge from the revised text; while for his

reign, at least, the manuscripts of this first recension differ even from each other. A reading of the standard

version suggests, in fact, that Barani may have drafted an account of 'Ala' al-Din's reign as a separate work

and subsequently incorporated it in a larger history, but without amending his treatment of the first few

months.2

Chronology is no less a problem for this period than for the previous century. Barani's very

attention to analysis at the expense of chronology raises difficulties for the student of Muhammad's reign in particular. Like 'Isami, he furnishes few dates, although there are more in his first recension than in the

revised text; and in the latter Barani expressly denies that he is presenting the crises of Muhammad's reign

in strict chronological order.3 It is with some difficulty that a narrative framework can be reconstructed by

1 Here I have relied primarily on Bodleian ms. Elliot 353 and a ms. in the private collection of Mr

Simon Digby; but certain readings have been checked against RRL Persian ms. 2053.

2 The events from late 695 to the autumn of 696 are thus covered twice: TFS, 239, 242-6. 3 Ibid., 468, 478.'

151

~ means of these accounts with some assistance from Ibn Battuta, although he too is sparing of

dates and many episodes to which he alludes preceded his arrival in 734/1333.4

How much is lost, we cannot be sure. Barani tells us that 'Ala' al-Din Khalji's reign was

distinguished by a number of prominent historians (mu'arrikhan). Kabir al-Din, the sultan's amir-i dad-i

lashgar, is said to have excelled in the skills of a secretary (dabiri) and in composition (inshd'). He

allegedly completed volumes of fath-ndmas and also wrote a Ta'rikh-i 'Ald'i. Amir Arslan *Kalahi, too, had

such a prodigious memory for the deeds of past sultans that he was able to answer 'Ala' al-Din's questions

without recourse to books. Barani, claiming merely that his own Ta'rikh-i Firuz-Shahi was based simply on abridgement of past histories, does not say explicitly that these included anything written by Kabir al-Din

or Amir Arslan.5 In any event, no work by either man has survived, and what we are told of Amir Arslan

does not suggest that he wrote a history of his own.

Of the written sources known definitely to have existed, some would have been invaluable:

Bijapuri's Mulhaqdt, for instance, the 'long' qasida on Firuz Shah's exploits composed by Mutahhar, and the

Shdh-Ndma composed by Muhammad b. Tughluq's court poet, Badr-i Chach, and described by Bada'uni as

a 'treasure'.6 'Afif, who dubs himself the author of 'the histories of sultans',7 claims to have written

biographies (mandqib) of 'Ala' al-Din, of Ghiyath al-Din Tughluq, of Muhammad b. Tughluq and of

Muhammad b. Firuz, and an account of Temur's sack of Delhi in 801/1398 (dhikr-i khardbl-yi DihlJ);8 but

no trace of these works exists today. Similarly lost is another Ta'rlkh-i Flruz-Shdhi, to which a certain cAbd

al-cAziz[-i?] Shams *Bhanawri lays claim in the preface to his translation of the Kitdb-i Bardhi.9 In the 1540s a certain Husam Khan composed in Gujarat a Tabaqdt (or Ta'rikh)-i Bahddur-ShdhJ which is no

longer extant but was

4 The difficulties are compounded by later writers, beginning with TMS, where the sultan's

campaign to Nagarkot in 738/1337 is wrongly identified with an earlier expedition sent to the Qarachll

region; this misled Ishwari Prasad, A history of the Qaraunah Turks in India (Allahabad, 1936), 126ff. The

older schema based on this false chronology from 739 onwards is thereby skewed: see Sir Wolseley Haig,

'Five questions in the history of the Tughluq dynasty of Dihli', JRAS (1922), 336-65; and for some of the

problems, N. Venkata Ramanayya, 'The date of the rebellions of Tilang and Kampila against Sultan

Muhammad bin Tughlaq', Indian Culture 5 (1938-9), 135-46, 261-9 (though his date for Fakhr al-Din's

revolt in Bengal, ibid., 138, 140, is surely too late). 5 TFS, 14, 361. For Taj al-Din Iraqi, see ibid, 358.

6 Mutahhar: Bihamadkhani, fols. 407a, 413a-414b (tr. Zaki, 4, 15). Badr: Bada'uni, I, 241; P.

Jackson, 'Badr-i CacT, Enc.Isl2, Supplement; the date of completion of the Shah-Nama is given in a

chronogram in Badr's Qasa'id, ed. M. Hadi cAli (Kanpiir, n.d.), 85 (see also ED, III, 572-3).

7 ‘Afif, 256.

8 ‘Ala' al-Din: ibid, 478. Ghiyath al-Din: ibid, 27, 36. Muhammad b. Tughluq: ibid, 42, 51, 92,

274, 394, 451. Muhammad b. Firuz: ibid, 148-9. 273, 428, 440. Sack of Delhi: ibid., 182, 185.

9 IOL Persian ms. 1262, fol. 2b; Ethe, Catalogue, col. 1112 (no. 1997), assumes that the author is

to be identified with cAfif.

~ utilized by Ulughkhanl and is cited by Firishta.10 From the excerpts we have, it is clear that

Husam Khan had access to sources other than Sirhindi or Bihamadkhani; his chronology, however, bears

marked similarities to Sirhindi's.

The one chronicle contemporary with 'Ala' al-Din that has come down to us is Amir Khusraw's

prose work, Khaza'in al-Futuh, which was completed in 711/1311-12 and provides a florid and bombastic

account of the sultan's victories over various Mongol attacks and of Kafur's campaigns in the south; the

poet himself calls it a 'fath-nama'.11 Khusraw's last historical works are the Nuh Sipihr, written in 718/1318 under Qutb al-DIn Mubarak Shah and incorporating an account of the sultan's expedition to the Deccan; the

Diwal Rani, which was completed just after the end of the Khalji period, in 720/1320, and contains details

not found elsewhere; and the Tughluq-Ndma, commemorating the overthrow of the usurper Nasir al-DIn

Khusraw Shah and the accession of Ghiyath al-DIn Tughluq (720-724/ 1320-1324).12

In addition, although

the treatise on prose composition, Rasa'il al-I'jaz, which Khusraw produced in 719/1319-20, is rather

suspect, some of the documents it contains appear to be based in part on genuine originals.13

For the Tughluqid era, although it seems that we must discount the fragment of the alleged

memoirs of Muhammad b. Tughluq, which the majority of scholars no longer regard as authentic,14

we

have access to richer and more varied material than for any of the previous dynasties. There are a few works composed in order to commemorate specific events, like the Basatin al-Uns (726/1325-6) in which

Taj al-Din Muhammad-i Sadr-i A'la Ahmad-i Hasan 'AydawsI, known as Ikhtisan-i Dabir, describes

Tughluq's Lakhnawtl campaign of 724/1324,15 and some of the verses of Muhammad's court poet Badr-i

Chach. The extensive correspondence (insha') of 'Ayn al-Mulk Ibn Mahru, who served both Muhammad

and Firuz Shah as governor of Multan and who died at some point before

10 On Husam Khan, see the introduction to AHG, Ross's edition of the Zafar al-Walih, II,

xxvii-xxix. He is cited by Firishta, I, 3, and II, 512. 1' KF, 170; see ibid., 26, for the current year.

12 Useful summary of contents in Syed Hashimi, 'The Tughluq-namah', IC 8 (1934), 301-12, 413-

24. 13 For an analysis, see S. H. Askari, 'Risail [sic]-ul-Ijaz of Amir Khusrau: an appraisal', in Dr.

Zakir Husain presentation volume (Delhi, 1968), 116-37; idem, 'Material of historical interest in I'jaz-i-

Khusravi', MIM 1 (1969), 1-20.

14 BL ms. Add. 25785 (of TN), fols. 316ff.; tr. in A. M. Husain, Tughluq dynasty (Calcutta, 1963),

265-76, and facsimile of text at end. See ibid., 567-72, for an analysis of the document, which Husain

believed to be authentic; for the contrary (and now widely accepted) view, Nizami, Studies in medieval

Indian history, 76-85, and his On history and historians, 198-205; the arguments are reviewed in Stephan

Conermann, Die Beschreibung Indiens in der 'Rihla des Ibn Battuta, IU, 165 (Berlin, 1993), 47-9.

15 S. H. Askari, 'Historical value of Basatin-al-Uns', JBORS 48, part 2 (1962), 1-29. Ikhtisan,

Basatin, BL ms. Add. 7717, fol. 19b, gives the current year as 726.

~ 772/1370,16 contains a good deal of material on fiscal and military affairs, mostly relating to

Firuz Shah's era, although some letters date from the time of Muhammad. In the Dastur al-Albab which

Hajji 'Abd al-Hamid Ghaznawi began in 734/1333-4 and completed in 766/1364-5, we have a treatise on

the administration of the Sultanate from the pen of a clerk (muharrir), of particular value on the subject of

taxation.17 For the four-teenth century, lastly, we also possess material relating to the sufi orders (silsilas,

tariqas), in which reference is sometimes made to contemporary political events. Chief among these, for

our purposes, are the collected biographies of sufi shaykhs, the Siyar al-Awliya', of Muhammad b. Mubarak

KirmanI (Amir Khwurd; d. 770/1368-9); Amir Hasan Dihlawi's Fawa'id al-Fu'ad, comprising the

discourses (malfuzat) of the influential Chishti Shaykh Nizam al-Din Awliya' (d. 725/1325); and Hamid Qalandar's Khayr al-Majalis (c. 755/1354), which contains those of Shaykh Nasir al-Din Mahmud

Chirdgh-i Dihli ('the Lamp of Delhi').18

Turning to sources from outside India, the so-called correspondence (mukdtibai) of the Ilkhanid

statesman Rashid al-DIn must be discounted as a contemporary source for Indian affairs: many of the

letters undeniably reflect a considerable familiarity with the administrative machinery in Persia and with

the nature of the India trade, but it seems difficult to avoid the conclusion of Reuben Levy that they

emanate from India itself in the fifteenth century.19 With the onset of the Tughluqid era, data from external

sources become more plentiful. Wassaf's latest information on the Sultanate relates to the accession of

Muhammad b. Tughluq;20 but the opening up of Muslim India during Muhammad's reign to diplomatic

contact with distant parts of the Islamic world, especially Mamluk Egypt, is reflected in the encyclopaedia Masdlik al-Absar of al-'Umari (d. 749/1349), who was able to amass a veritable dossier of information

about India; in notices on Muhammad and his empire in the biographical dictionaries - al- Wafi bi'l-

Wafayat and Ayan al-'Asr - of al-Safadi (d. 764/1363), the chronicles of Ibn

16 His death is mentioned in SFS, 154.

17 The author states that he was forty-four in 734 and that he was 917 months old, i.e. in his

seventy-seventh year, on completing the book. This is said to have occurred in 760: DA, fols. 3a, 4b, cited

without question by Rashid, 'Dastur-ul-Albab', 59. But the details contradict one another, and the correct

year must be 766. 18 For these and other (often spurious) works, see Mohammad Habib, 'Chishti mystic records of

the Sultanate period', MIQ, 1 (1950), no. 2, 1-42; Nizami, On history and historians, 163-80.

19 R. Levy, 'The letters of Rashid al-Din Fadl-Allah', JRAS (1946), 74-8. See now A. H. Morton,

'The letters of Rashid al-DIn: Ilkhanid fact or Timurid fantasy?', in Amitai-Preiss and Morgan, The Mongol

empire and its legacy (forthcoming). K. A. Nizami, 'Rashid-u'd-Din Fazlullah as an Ilkhanid envoy to the

court of Ala-u'd-Din Khalji', PIHC 29 (Patiala 1967) (Patna, 1968, 2 vols.), I, 139-43, and in On history

and historians, 99-104, acknowledges some of the problems, but is inclined to accept the authenticity of the

letters.

20 Incorrectly placed in 723/1323. But the current year at one point appears as 727 (Wassaf, 607).

The date 718 (ibid, 608) is manifestly an error for 728: Barthold, Turkestan, 49 n.2.

~ Abi'l-Fada'il (fl. 1340) and Shabankara'i (738/1337-8),21 and the travel narrative of Ibn Battuta.

Of these, the Tuhfat al-Nuzzar (often called simply the Rihla) of Ibn Battuta, who spent several years in the

Sultanate, furnishes a picture of life at Muhammad's court and in his dominions between 734/ 1333 and c.

748/1347 that in its vividness is unmatched elsewhere. The archive of the shaykhs of Jam in Khurasan,

found in the fifteenth-century Fara'id-i Ghiyathl of Yusuf b. Muhammad b. Shihab al-Jami (Yusuf-i Ahl),

contains correspondence with the Delhi government during the reigns of both Muhammad and his

successor.22

Sources from 752/1351 onwards

The earlier recension of Barani's Ta'rikh-i Flruz-Shdhi embraced only the first four years of FIruz

Shah's reign, and the revised text was completed two years after that. We are told that when the sultan

desired a history of his reign to be written, and invited applications from would-be chroniclers following

Baranl's death in c. 762/1360-1, none came forward, and he was reduced to composing his own account,

which he caused to be carved on the dome of the Jami' Masjid at his new capital, Firuzabad.23 Fortunately,

however, this dearth of historiographical enterprise did not last, and for FIruz Shah's reign (752-90/1351-

88) we have access to a crop of literary sources. What might be called 'official' history is represented by the

copy of FIruz Shah's lengthy inscription that has come down to us as Futuhdt-i Firiiz-Shdhi,24 and by the

panegyrical SJrat-i Flruz-Shdhi, produced for the sultan soon after c. 772/1370 by an anonymous author

who may have been the poet Mutahhar.25 It was not, however, until the early fifteenth century that 'Afif

wrote his Ta'rikh-i Flruz-Shdhi, which is the fullest source for the reign. 'Afif, who belonged to a bureaucratic family that had served the Tughluqids and himself worked in the diwan-i wizarat in the middle

of the 1380s,26 intended his biography of the sultan to be a sequel to Barani's work, comprising the ninety

muqaddimas which the older historian had

21 D. P. Little, 'Al-Safadi as a biographer of his contemporaries', in Little (ed.), Essays on Islamic

civilization presented to Niyazi Berkes (Leiden, 1976), 190-210. C. E. Bosworth and P. Jackson,

'Shabankara'i', Enc.Isl2.

22 On this work, see Jean Aubin, 'Le khanat de Cagatai et le Khorassan (1334-1380)', furcica 8

(1976), 20 n.19; PL, III, part 2, 251-2 (no. 428).

23 'Afif, 176-7: around the time of the sultan's return from his Jajnagar campaign, which occurred

in Sha'ban 762/June-July 1361 according to SFS, 14, and in Rajab/May-June according to TMS, 130.

Hodivala, Studies, I, 129, warns against taking this as a precise indication of the date of Barani's death.

24 K. A. Nizami, 'The Futuhat-i-Firuz Shahi as a medieval inscription', in PSMI, 28-33, and in his

On history and historians, 205-10.

25 As suggested by K. A. Nizami, Supplement to Elliot and Dowson's History of India, III (Delhi,

1981), 63.

26 'Afif, 487-8: the context is the disgrace of Shams al-Din Abu-Rija, which occurred in 785/ 1383-

4 (ibid., 497-8). For 'Afif's forebears in the Tughluqids' service, see ibid., 37, 127, 130-1, 138, 145, 196,

197,339.

~ announced his intention of writing but had not lived to complete.27 Regrettably, the text we have

is defective at the end, to judge from the list of contents supplied by 'Afif himself. Both Sirhindi (838/1434)

and Bihamad-khani (842/1438) supply information on Firuz Shah's reign which is not found in 'Afif 's

Ta'rikh; and they continue to be our principal sources from the 1380s onwards. Bihamadkhani's general

chronicle, the Ta'rikh-i Muhammad!,28 which down to 755/1354 relies on Barani's earlier recension,

becomes at that juncture an original source; and though less detailed than Sirhindi's work it has the

particular merit that it was composed not at Delhi but in the newly autonomous principality of Kalpi and

hence provides us with a different vantage-point from that of earlier chroniclers.

Temiir's invasion of India is of course covered in some detail by the Timurid chronicles. The most

immediately contemporary of these was the lost Ruz-Nama-yi Futuhdt-i Hindustan of Qadi Nasir al-Din

'Umar, who accompanied the conqueror. This was abridged both by Ghiyath al-Din 'All Yazdi, whose Ruz-

Nama-yi Ghazawat-i Hindustan has survived, and by Nizam-i Shami, who incorporated it into his Zafar-

Ndma, an account of Temur's career completed in 806/1404. Both works (and possibly also Qadi Nasir al-

Din's original text) were in turn utilized by Sharaf al-DIn 'Ali Yazdi when he came to produce his own

Zafar-Ndma in 828/1424-5.29

The Khalji Sultans Barani depicts 'Ala' al-Din Khaljl as an unlettered soldier with little time for the 'ulama', but a man

of boundless ambition who had to be dissuaded from founding his own religion; he was amazed that such a

sultan, who set realpolitik above the injunctions of the Shari'a, could have prospered to the extent that he

did.30 Yet for 'Isami the contrast could not have been stronger between 'Ala' al-Din, who had done so much

to implant Islam in India, and the contemporary sultan, Muhammad b. Tughluq, who had presided over its

collapse;31 and at the time of Ibn Battuta's visit to Delhi a few years before, the citizens evidently looked

back on 'Ala' al-Din's era as a golden age.32 It is true that the reign was marked both by the repulse of

formidable Mongol invasions and by spectacular advances at the expense of independent Hindu powers in

Rajasthan and the south. The capacity of the Sultanate to raise large and effective military forces was

placed on a new footing by means of economic reforms which kept prices low in the capital.

Ibid., 29-30; cf. TFS, 529-30, 602.

On which see Peter Hardy, 'The Tarikh-i-Muhammadi by Muhammad Bihamad Khani', in

Gupta (ed.), Essays presented to Sir Jadunath Sarkar, 181-90.

For the relationship between these works, see John E. Woods, 'The rise of Timurid

historiography', Journal of Near Eastern Studies 46, part 2 (1987), 93-5.

TFS, 261-6, 289. For Barani's view of the sultan, see generally Hardy, Historians, 32-4.

FS, 604, 605-6 (tr. 898, 900-1). 32 IB, III, 184 (tr. Gibb, 640).

~ Barani describes such achievements in terms of the miraculous.33 But he also cites the opinion of

the mystic Shaykh Bashir that 'Ala' al-Din's regime, founded as it was on his uncle's murder, was inherently

unstable;34 and in the chronicler's own eyes the fate of 'Ala' al-Din's sons was retribution for Jalal al-Din's

murder.35

During 'Ala' al-Din's final illness, which BaranI calls dropsy (istisqa), his heir, Khidr Khan, was

imprisoned in Gwaliyor at the instigation of the sultan's na'ib, the slave Kafur, and shortly blinded

following his father's death on 7 Shawwal 715/4 January 1316;36 his brother ShadI Khan suffered the same fate. Kafur, whose aim, if BaranI is to be trusted, was to destroy the entire Khalji dynasty,37 ruled through

an infant son of 'Ala' al-Din, Shihab al-DIn 'Umar; but he enjoyed power for a mere thirty-five days before

being murdered by 'Ala' al-Din's paiks. Another son then assumed the regency, but soon displaced the child

ruler, on the pretext that the boy's mother had tried to poison him, and himself reigned as Qutb al-Disn

Mubarak Shah (716-20/1316-20). In 718/1318, on the return march from a campaign in the Deccan, Qutb

al-Din's cousins, the descendants of Jalal al-Din's brother Khamush (Yughrush Khan), were executed on

suspicion of complicity in a plot to assassinate the sultan;38 Khidr Khan, Shadi Khan and 'Umar were put to

death; and their remaining brothers were despatched to Gwaliyor. In Jumada II 720/July 1320 the sultan's

favourite, the Indian slave Hasan, entitled Khusraw Khan, had him murdered and ascended the throne as

Nasir al-DIn Khusraw Shah - the only Delhi monarch, in fact, who was an Indian convert to Islam. All 'Ala'

al-Din's surviving sons were now massacred.39 The Khaljl dynasty appears to have been completely

exterminated. When one of 'Ala' al-Din's senior lieutenants, Ghazi Malik Tughluq, the muqta' of Deopalpur, posing as the avenger of his master's heirs, marched on Delhi and overthrew Khusraw Shah, no

member of the dynasty could be found to take the throne.40 Ghazi Malik himself was accordingly

proclaimed as Sultan Ghiyath al-Din Tughluq Shah (720-724/ 1320-1324).41

For some, says BaranI, the parallel between the reigns of Mu'izz al-Din Kayqubad and Qutb al-

Din Mubarak Shah Khaljl was striking.42 In both cases a young and profligate ruler succeeded a harsh and

despotic one,

33 TFS, 339.

34 Ibid., 377-8. 35 Ibid., 237.

36 The date for 'Ala' al-Din's death supplied in DR, 259. TMS, 81, gives 6 Shawwal, and TFS, 369,

the evening of that day; FS, 344 (tr. 524), has 11 Shawwal.

37 TFS, 375.

38 Ibid., 393. FS, 363-4 (tr. 562-3).

39 DR, 273-85; Tughluq-Nama, 23-4, 31-2, 47; TFS1 Bodleian ms, fol. 172a/Digby Coll. ms., fol.

146b (not in TFS, 408). For a relatively detailed version of their fate, as current in Delhi some years later, see IB, III, 189-90, 191-4 (tr. Gibb, 643, 644-5), who believed, however, that all Qutb al-Din's brothers

were put to death during his reign.

40 TFS, 421-2; and see also 237. Tughluq-Nama, 140-1, does not actually confirm that the Khalji

dynasty was extinct.

41 For the date of Tughluq's death, usually given as 725/1325, see appendix V.

42 TFS, 383, 387-8.

~ leading to a general relaxation of state authority and public morals. Yet Qutb al-DIn manifested greater military energy than his precursor, heading a campaign which reasserted imperial rule over the

Deccan in 718/1318. For a time at least the young sultan won great popularity through the abrogation of his

father's repressive measures. Many matters are unexplained however, and Barani is guilty of his customary

inconsistency. Even allowing for hyperbole, it is not clear, for instance, why, if the sultan could not bear to

be parted even for one hour from Khusraw Khan,43 he was prepared to send him on a lengthy expedition to

the far south. Nor do the chroniclers indicate why amirs who threatened to report the favourite's treasonable

plans to the sultan during that campaign ranged themselves under his banner against Tughluq a few years

later (pp. 177, 179 below) - especially since the latter's revolt is portrayed by both Amir Khusraw and

Barani as a Holy War (ghaza').44

One answer to this second problem may well be that Nasir al-Din Khusraw Shah's rule was less

repugnant than our sources would have us believe. Barani is conceivably right when he alleges that idolatry

was practised within the royal palace, presumably by those of Khusraw Shah's adherents who were not

converts. But his story, on the other hand, that Khusraw Shah and his lieutenants treated Qur'ans with

blatant disrespect and set up idols in mosques is hardly worthy of credence; it is noteworthy that the

Tughluq-Nama talks of idolatry in less specific terms and that the version of events heard by Ibn Battuta,

who singles out for mention only a prohibition on slaughtering cows, is rather less extreme.45 Yet even if

Khusraw Shah's regime cannot be regarded as anti-Muslim, it is still necessary, on the other hand, to

explain the widespread acquiescence in the murder of Qutb al-Din Mubarak Shah. Possibly Qutb al-Din's

assumption of the caliphal title, which is not mentioned in the literary sources but which can be dated to

717/1317-18,46 had scandalized many Muslims. Some hint may be found, too, in Barani's claim that Qutb

al-Din had been on bad terms with the Chishtl shaykh Nizam al-Din Awliya' as a result of the murder of Khidr Khan.47 Given the sultan's poor relations with the Chishtl khanaqah, it is conceivable that in the eyes

of the shaykh and his

43 Ibid., 382.

44 Ibid., 399-400, for the amirs' threats. Tughluq-Nama, 62, 100, and TFS, 415-16, for holy war.

45 Ibid., 410-11. Tughluq-Nama, AA. IB, III, 200 (tr. Gibb, 648). For a balanced assessment of

Khusraw Shah's reign, see Lal, History of the Khaljis, 313-16; also Hardy, 'Force and violence', 172.

46 He is called merely Qasim Amir al-Mu'minin in an inscription of 5 Muharram 718/9 March

1318: Z. A. Desai, 'The Jalor 'Idgah inscription of Qutbu'd-Din Mubarak Shah Khalji', EIAPS (1972), 12-

19. The title of caliph is found in inscriptions later in that year: Yazdani, "Inscriptions of the Khalji Sultans

of Delhi', 38-40; Z. A. Desai, 'Khalji and Tughluq inscriptions from Gujarat', ElAPS (1962), 4-5. But it

appears on coins of 717: CMSD, 96-101. Qutb al-DIn is frequently addressed as khalifa in NS.

47 TFS, 394.

~ sympathizers the Indian upstart was preferable to the Khalji. Qutb al-Din may thus have

alienated support and played into the hands of Khusraw Khan and his party.

Shaykhs and chroniclers All monarchs and their kingdoms, wrote 'IsamI, lay under the protection of a saint; and the first

step of Providence when it wished to destroy a country was to effect the saint's departure.48 Thus for him

the death of Nizam al-Din ushered in the horrors endured by Delhi in the era of Muhammad b. Tughluq,49

and the prosperity of Dawlatabad, prior to the revolt against Muhammad from 745/1344 onwards, could be

attributed to the presence of two shaykhs, Burhan al-DIn and Zayn al-DIn.50 Sufis from Khurasan had been

present in India since the Ghurid era, and two orders had grown up -the Suhrawardiyya, with their principal

base at Multan, and the Chishtiyya, whose headquarters were in Delhi. The orders differed in their attitudes

towards the state: for the Suhrawardiyya, association with the powerful was permitted; the Chishti shaykhs,

by contrast, eschewed contact with the court and the nobility and rejected revenues and government service

(shughl). Relations between the two groups were nevertheless harmonious and based on mutual respect.51

The view expressed by 'IsamI is especially common, of course, among those who recorded the

discourse of shaykhs and the hagiographers like Kirmani (Amir Khwurd), who saw the very presence of

Muslims in India as a miracle (karamat) on the part of the Chishtl Shaykh Mu'ln al-Din;52 Amir Hasan

Dihlawi thought that Multan had been saved from the Mongols in Qubacha's time through the intervention

of Shaykh Qutb al-Din Bakhtiyar Kaki.53 But such convictions were shared by other writers whose

lifestyles and fortunes were less closely bound up with the orders. 'Afif believed it was the shrine of Qutb

al-Din Munawwar that preserved Hansi during Temur's invasion.54 Recalling to mind the tyranny of 'Ala'

al-Din KhaljI, Barani could conceive of no reason for the continued success of the sultan's regime other

than the fact that Nizam al-Din Awliya' graced his capital.55

The numinous power or spiritual charisma (baraka) of a shaykh could be seen as territorial and as

constituting a rival locus of authority (wilayat) to

48 FS, 455-6 (tr. 687-8).

49 Ibid., 456-7 (tr. 688-9).

50 Ibid., 458-9, 461-2 (tr. 691-2, 696-7).

51 K. A. Nizami, 'Early Indo-Muslim mystics and their attitude towards the state', IC 22 (1948),

388-92, 395-8, and 23 (1949), 13-21. Aziz Ahmad, 'The sufi and the sultan in pre-Mughal Muslim India',

Der Islam 38 (1962), 142-7. S. Digby, 'The sufi shaikh as a source of authority in mediaeval India', in Gaborieau (ed.), Islam et societe, 63-5.

52 Siyar, 47, cited in Digby, 'The sufi shaikh as a source of authority', 72.

53 Amir Hasan Dihlawi, Fawa'id al-Fu'ad, 185.

54 ‘Afif, 82; and cf. also 133, where it is attributed to the baraka of Munawwar's successor.

55 TFS, 324-5.

~ that of the sultan.56 Several anecdotes show shaykhs conferring the sover-eignty on a prince. Stories were current in Juzjani's day that kingship had been bestowed by faqirs both on Husam al-Din 'Iwad

and on Iltutmish; similar tales are told regarding Balaban and 'Ala" al-Din Khalji; and 'Afif reports no less

than four anecdotes in which Firuz Shah is promised the crown by shaykhs, among them Nizam al-Din

Awliya'.57 That the shaykh's khanaqah might also serve as a rallying-point for disaffected elements had

been thrown into relief by the Sidi Muwallih affair in the time of Jalal al-Din Khalji (above, p. 83).

In these circumstances, relations between court and khanaqah might not always be harmonious,58

and for our chroniclers one of the most important criteria in evaluating a sultan's reign was his treatment of

holy men. Here, for all his faults, 'Ala' al-Din, who demonstrated a growing attachment to Shaykh Nizam

al-Din Awliya' during his last years, proved relatively sound.59 The reign of his son Qutb al-DIn Mubarak

Shah, however, was vitiated by his relations with Nizam al-DIn. When the saint condemned the murder of

Khidr Khan, who had been his disciple (murid), the sultan responded with slights and threats and attempted to set up the immigrant Shaykhzada Shihab al-DIn JamI and the Suhrawardi Rukn al-Din of Multan as his

rivals in Delhi.60 Nizam al-DIn was extremely influential: we are told of several notables who were among

his disciples.61 During the brief reign of Nasir al-Din Khusraw Shah, Nizam al-Din accepted gifts of money

from the usurper, and spent them on charitable causes. He thus made a new enemy of Ghiyath al-Din

Tughluq Shah when that monarch sought to retrieve the sums disbursed by his predecessor.62 Hostility

between the two men persisted: Tughluq is said to have been contemplating further action against Nizam

al-DIn during the return march from Bengal just before his death; though the shaykh's ironic comment,

Dilli az tu dur ast ('Delhi is some way off for you'), is not reported by any author prior to Sirhindi.63

Nizam al-DIn survived only a few months into the reign of Muhammad b. Tughluq, with whom

his relations had been cordial: Ibn Battuta was told 56 Digby, 'The sufi shaikh as a source of authority', 62-3; idem, 'The sufi shaykh and the sultan: a

conflict of claims to authority in medieval India', Iran 28 (1990), 71-81.

57 Ibid., 75-8. See further Eaton, Rise of Islam, 82-6.

58 Nizami, 'Early Indo-Muslim mystics', IC 23 (1949), 312-21; for good relations, ibid., 165-70.

59

TFS, 332.

60 Ibid., 394, 396. For Qutb al-Din's close relations with Shaykhzada Jami, see IB, III, 294 (tr.

Gibb, 697); also Digby, 'The sufi shaykh and the sultan', 79 n.20. Rukn al-Din's own relations with Nizam

a]-DIn remained harmonious: idem, The sufi shaikh as a source of authority', 64.

61 'Afif, 69, 445; and cf. also TFS, 396.

62 For the cancellation of grants made from the treasury by Khusraw Shah, see ibid., 439.

63 TMS, 96-7: the remark is embellished in later sources. For more details, see Digby, 'The sufi

shaykh and the sultan', 72-4.

~ that Muhammad carried the shaykh's bier.64 Their contacts seem to have contributed to the strain between Tughluq and his heir-apparent. Yet the new sultan's own relations with shaykhs proved

problematic when he sought to recruit the talents of sufi shaykhs for service to the state.65 This created no

difficulty for the Suhrawardi order, which had never objected to involvement in the world's affairs: Mu'izz

al-Din, son of the SuhrawardI Shaykh 'Ala' al-Din Ajudhani, seems to have accepted the governorship of

Gujarat without demur.66 The sultan's relations with the descendants of the Chishti shaykh Hamid al-DIn at

Nagawr were also cordial.67 But for most of the Chishtiyya, his policy constituted a major crisis. Ibn

Battuta retails numerous anecdotes demonstrating the shaykhs' resistance and the harsh punishments they

suffered in consequence. It was Amir Khwurd's opinion that Muhammad's dismal end far from the capital

was due to his treatment of holy men, chiefly Nizam al-Din's successor (khalifa), Nasir al-Din Mahmud

Chiragh-i Dihli.68

From Ghiyath al-Din to Firuz Shah

The Tughluqids (720-815/1320-1412) proved to be the longest-lived of the dynasties that ruled

over the Sultanate. During Tughluq Shah's brief reign, Bengal was again subjected to the sultan's

overlordship, the Kakatiya kingdom of Arangal (Tilang; Telingana) was annexed, and Muslim authority

was established over much of the Pandya kingdom of Ma'bar. Barani's view of Ghiyath al-Din Tughluq is

somewhat one-sided. He chooses to ignore the sultan's strained relations with Nizam al-Din, and praises

Tughluq for being in many respects a model Muslim ruler. He was like a father to his troops; his dominions

enjoyed justice and security; his piety and personal morality were above criticism. Tughluq is said to have

accomplished what 'Ala' al-DIn had done, but without bloodshed.69 But when he comes to describe the

punishment in Delhi in 721/1321-2 of those who had mutinied during the campaign in Tilang, Barani lets

his guard drop, revealing that the wives and children of the ringleaders were put to death.70 Yet there is no hint of condemnation here for a practice which had begun under 'Ala' al-Din and which Barani clearly

deplored.71

Tughluq perished when a newly constructed building at Afghanpur collapsed on him. Although

BaranI makes no such accusation, the suspicion

64 IB, III, 211 (tr. Gibb, 653-4); and see also MA, ed. Spies, 20 (German tr. 46)/ed. Fariq, 38 (tr.

Siddiqi and Ahmad, 45).

65 IB, III, 294 (tr. Gibb, 697).

66 TFS, 507, 512. SFS, 20-1 (tr. Basu, in JBORS 23 [1937], 98). Siyar, 196.

67 K. A. Nizami, 'Some documents of Sultan Muhammad bin Tughluq', MIM 1 (1969), 305-6,

307,309-13.

68 Siyar, 245-6, cited in Digby, 'The sufi shaykh and the sultan', 74. For Muhammad and the

Chishtiyya, see generally Nizami, 'Early Indo-Muslim mystics', IC 24 (1950), 60-5.

69

TFS, 445.

70 Ibid., 449.

71 Ibid., 253.

~ that his eldest son and designated heir, Muhammad (Ulugh Khan), had contrived his death was

shared by 'Isami and by al-Safadi's informants, while Ibn Battuta attributes it to the skill of the intendant of

buildings (shihna-yi imarat) Ahmad b. Ayaz, whom Muhammad rewarded with the post of wazir.72 The

smooth transition that followed Tughluq's death might have seemed to reinforce the impression that a new

era of stability had dawned, for Muhammad was apparently the first sultan to enjoy a peaceful succession.

The image of the sultan conveyed by foreign writers and fostered by his own propaganda is one of a mighty

warrior for the cause of Islam, whose triumphs are unprecedented and who unlike his predecessors has

cowed the Mongols.73 But in the event the reign of Muhammad b. Tughluq (724-52/1324-51) was characterized by rebellion and disaster. Although a number of revolts in the years 727-8/1326-8 were

suppressed, the sultan embarked on various ambitious projects which entailed considerable expenditure.

The effects of their failure were accentuated by plague and famine. A further wave of rebellions from

734/1334 onwards absorbed the attention of Muhammad and his lieutenants, and led to the definitive loss

of Ma'bar, Tilang and Bengal; while a new Hindu power emerged from c. 1336 at Vijayanagara. Although

he secured a temporary respite after 741/1340, and successfully applied to the 'Abbasid Caliph at Cairo for

a diploma of investiture in 744/1343, his last years witnessed a widespread revolt by members of the

military class, the amiran-i sada ('amirs of a hundred') in Deccan and Gujarat. The rebels in Gujarat were

defeated; but at Deogir (Dawlatabad) in 748/1347 the rebel leader Hasan Gangu, the founder of the

Bahmanid dynasty, established an independent sultanate. When Muhammad died near Thatta on 21

Muharram 752/20 March 1351 he wielded no authority south of the Vindhyas.

Muhammad b. Tughluq posed a problem for the historians: even the unimaginative Sirhindl

interrupted his annalistic catalogue of events to try to explain the causes of the sultan's failure.74 Yet we

simply cannot take at face value all the charges levelled at the sultan by our principal sources. On certain

heads, their testimony overlaps; to a degree they paint a similar picture of Muhammad's character. Barani,

'Isami and Ibn Battuta all comment, for instance, on the sultan's interest in philosophy;75 but that is

72 FS, 420 (tr. 633). al-Safadi, Wafi, III, 172, partial tr. M. S. Khan, 'An undiscovered Arabic

source of the history'of Sultan Muhammad bin Tughlaq', IC 53 (1979), 187. IB, III, 212-15 (tr. Gibb, 654-

6).

73 Conquests and spread of Islam: Shabankara'i, 87-8, 287; MA, ed. Spies, 29 (German tr. 55)/ ed.

Fariq, 53 (tr. Siddiqi and Ahmad, 54). Mongols: MA, ed. Lech, 40 (German tr. 118). See Muhammad's own

appeal to the notables of Transoxiana in FG, SK ms. Fatih 4012, fol. 456b.

74 TMS, 113-15.

75 TFS, 464-5. FS, 510 (tr. 759). IB, IV, 343 (tr. Gibb and Beckingham, 929). For Barani's hostility

towards philosophy, see FJ, 16, 168-9; tr. in W. T. de Bary, Sources of Indian tradition (New York, 1958),

481 -2.

~ not to say that they comprehended it. Professor Nizami has argued persuasively that Muhammad

was greatly influenced by the Syrian scholar and jurist Ibn Taymiyya (d. 728/1327), whose pupil 'Abd al-'Aziz Ardabili received a warm welcome at Muhammad's court.76 Ibn Taymiyya's aim was to reinvigorate

what he saw as decadent Islamic society. To this end, he sought to promote both ijtihad (fresh interpretation

of religious law) and jihad (holy war), and rejected the separation between state and religion as advocated

by the Chishtiyya among others. According to Nizami, Muham-mad's attested view that 'Religion and the

State are twins',77 his concern to enforce orthodox Islamic observance and practice, his attempts to press

sufis into the service of the state, and his adoption of the style al-Mujahidfi sabili'lldh (The Warrior in the

Path of God') were all symptomatic of his attachment to the ideology of Ibn Taymiyya; but his attitudes

were misunderstood by those, like Barani and 'Isami, who were unacquainted with currents of thought in

the wider Islamic world.78

BaranI, 'Isami and Ibn Battuta speak with one voice regarding Muham-mad's penchant for

inflicting harsh punishments.79 But whereas BaranI and Ibn Battuta, like al-'Umari's informants, are also

impressed by his gener-osity and by his concern for orthodoxy,80 'Isami - a hostile witness writing for a

rival monarch in the breakaway Bahmanid Sultanate of the Deccan -has nothing good to say of him

following an alleged change in Muham-mad's temperament two years into the reign.81 For 'Isami,

Muhammad is above all an apostate who consorts with Hindus and has thereby rendered it lawful for

orthodox Muslims to repudiate his authority and to take his life. His suspension of the Friday khutba

pending the arrival of a diploma from the caliph (the context supplied by BaranI) is distorted as the

abrogation of the requirements of Islamic worship.82 'Isami, of course, makes no mention whatever of the

caliphal diploma. In his account of the first check administered to the rebel forces of Nasir al-Din Isma'il

*Mukh in the Deccan, the insurgents are depicted as the 'faithful' (mu'minan) and Muhammad's army as the

forces of chaos (fitna)83 76 IB, II, 75-6, and III, 252-3 (tr. Gibb, 312-13, 676).

77 Siyar, 196.

78 K. A. Nizami, 'The impact of Ibn Taimiyya on South Asia', JIS 1 (1990), 120-34.

79 TFS, 459, 460. FS, 446, 468, 472 (tr. 675, 704, 708-9). IB, III, 216, 295-316 (tr. Gibb, 657, 695-

708), provides numerous examples.

80 Muhammad's generosity: TFS, 460-2; IB, III, 216, 217, 243ff. (tr. Gibb, 657, 658, 671ff.); MA,

ed. Spies, 21-5 (German tr. 47-51)/ed. Fariq, 41-7 (tr. Siddiqi and Ahmad, 46-50). His attention to

orthodoxy: TFS, 459, 460; IB, III, 216, 286-8 (tr. Gibb, 657, 693-4); MA, ed. Spies, 21, 25-6 (tr. 46-7, 51-

2)/ed. Fariq, 38-41, 47-8 (tr. Siddiqi and Ahmad, 45-6, 50-1).

81 FS, 424 (tr. 650).

82 Ibid., 515 (tr. 764-5); cf. also 450-1 (tr. 681-2), where Muhammad is compared unfavour-ably

with the epic tyrant Dahhak. For the suspension of the prayers, see TFS, 492.

83 FS, 535 (tr. 790); see also 538 (tr. 793), and 520 (tr. 771) for a description of some of

Muhammad's supporters as 'enemies of the Prophet's faith'. For a contrast between the two authors, see

Nizami, On history and historians, 133-4; Conermann, Beschreibung Indiens, 112-23.

~ Barani's attitude is more complex. Muhammad was 'the wonder of the age', who represented a

truly bewildering combination of the opposing qualities required in a sovereign (above, p. 54): in particular,

he (Muhammad) failed to distinguish between the duties of sultan and prophet.84 The differences between

the two recensions of his work are at their most glaring in their treatment of this reign, and Dr Hardy, in a

comparison of the two versions, has drawn attention to the fact that the second is more moralistic in tone

and attributes a greater degree of responsibility to the sultan.85 Dedicating his work to Muhammad's

successor, a trusted servant of the late monarch whose own reign nevertheless witnessed a reaction against

Muhammad's excesses, Barani is evidently anxious to distance himself from the previous regime. It seems

that his need to do so grew between the two versions of the Ta'rikh.86 For Barani, the most heinous feature

of Muhammad's government had been the slaughter of Muslims, and in particular the harsh punishments meted out to the 'ulama', shaykhs, sayyids, sufis, qalandars and members of the clerical and military

classes.87 But having been in attendance on Muhammad for over seventeen years as a boon companion

(nadim),88 Barani was himself implicated in these crimes. Thus he is at pains to express remorse at his own

fear of speaking out against his late master's policies or of offering Muhammad salutary advice.89

It is,

however, difficult to assess what use Barani made of his Fatawa-yi Jahandar as a vehicle for criticism of

the late sultan. The picture he draws of the tyrant Yazdagird, for instance, is in some (though by no means

all) respects reminiscent of Muhammad.90

For all its defects, Barani's Ta'rikh (particularly the later recension) operates on a far higher plane

than 'Isami's work. The gulf between the two men emerges clearly in their handling of the creation of a second capital at Dawlatabad in the Deccan and of other enterprises such as the adoption of the so-called

token currency and the ill-fated Qarachil expedition. 'Isami, whose aged grandfather had died soon after

leaving Delhi for the south in the original emigration, devotes considerable space to the enormity of the

457-60. See the comments of Hardy, Historians, 37, and 'Didactic historical writing', 49-51.

85 Ibid., 51-7.

86 Iqtidar Husain Siddiqui, 'Fresh light on Diya' al-Din Barani: the doyen of the Indo-Persian

historians of medieval India', IC63 (1989), 71-7.

87 TFS, 460, 472, 497, for Muslims in general; 459, 465-6 for the ‘ulama' etc. 88 In TFS', Bodleian ms., fol. 196a/Digby Coll. ms, fol. 163b, he describes himself as dar miyan-i

nudama'. TFS, 504, for the number of years; cf. also 466, 497, where he calls himself a muqarrab.

89 Ibid., 466-7, 497, 517, for his silence; see also the comments of I. Habib, 'Barani's theory', 102.

90 FJ, 264-6. Yazdagird bribes an invader to retire, rather as Muhammad, in one tradition, is said

to have bought off the Chaghadayid khan Tarmashirin (see p. 232, n.101); and thereafter, like Muhammad,

raises the land tax (kharaj) by one-fifth and one-tenth (yaki ba-panj-u yaki ba-dah) in order to recruit a

fresh army: see below, p. 262. But his fate (being torn to pieces by his resentful subjects) does not resemble

Muhammad's.

~ Dawlatabad project.91 He sees Muhammad's tyranny as a divine punishment for the readiness of

Delhi's citizens to tolerate heresy and religious innovation (bid'at); the death of the saint Nizam al-DIn

Awliya' (725/1325) leaves the city bereft of the protection of his spiritual power; and the token currency

and the Qarachil campaign become yet further means of victi-mizing the capital when Muhammad

perceives that the exodus of its leading families has not sufficiently crippled its prosperity.92 We have here

an echo of stories about the sultan's antipathy towards the people of the capital that were current when Ibn

Battuta visited Delhi a few years later.93 The idea that Muhammad b. Tughluq, like certain of his

predecessors (above, p. 59), regarded the citizens of Delhi with suspicion and hostility is not as out-landish

as it might first seem; though precisely why he may have done so is obscure. As the chief impulse behind

the establishment of the second capital, however, this is quite unconvincing. Barani is doubtless more

realistic in pointing to the geographical location of Dawlatabad, which made it ideally suited to be the centre of a considerably expanded Sultanate. This is a perspective found also in external sources; although

as Roy observed, another reason for the sultan's choice of Dawlatabad was the desire to implant Islam more

securely in the Deccan.94

In outline the analysis of Muhammad b. Tughluq's reign given in Baranis revised Ta'rikh is the

best thing we have, and it does provide a reasonably serviceable framework. An air of brilliance

conceivably hung over the early years of the reign, and deluded Muhammad, who enjoyed the strong

position of being the first designated heir to succeed his father as sultan of Delhi, into believing that

nothing lay beyond his capacities. Almost from the moment of his accession, the extensive tracts that now

owed obedience to him were subjected to a control of greater intensity than in the time of any of his

predecessors.95 Barani claims that had he reduced the whole world he would not have tolerated the least island or closet being exempt from his authority (a view faintly echoed by one of al-'Umari's informants,

who believed that only the islands and a mere span of coastline lay outside Muhammad's empire);96 and it

certainly seems that he was determined to impose uniformity upon his dominions. Unfortunately, his vision

proved impossible to realize, and his efforts to implement it led to the loss of a significant proportion of his

empire. But although Baranis insights

91 FS, 447-8 (tr. 677-8).

92

Ibid., 424, 446, 454-6, 459-60, 466, 468 (tr. 650-1, 675-6, 686-9, 693, 702, 704). On 'Isamis

perspective, see also HN, 507. 93 IB, III, 314-15 (tr. Gibb, 707-8).

94 TFS, 473-4; also TFS1 Bodleian ms., fols. 190b-191a/Digby Coll. ms, fols. 159b-160a. al-

Safadi, Wafi, III, 174 (tr. Khan, 188); see also al-Safadi, A'yan al-'Asr, SK ms. Asir Efendi 588, fol. 2a. N.

B. Roy, 'The transfer of capital from Delhi to Daulatabad', JIH 20 (1941), 159-80 (esp. 160-8).

95 TFS, 468, 469. TMS, 97-8.

96 TFS, 458. MA, ed. Spies, 5 (German tr. 23)/ed. Fariq, 11 (tr. Siddiqi and Ahmad, 29).

~ are not therefore to be dismissed out of hand, the emphasis laid in the second recension on the illusory character of Muhammad's enterprises is in fact highly tendentious, in that it plays down the

connections between them (see chapter 13).

On Muhammad's death in 752/1351, the army commanders and other leading figures present in

Sind prevailed upon the late sultan's cousin and amir-hajib, Firtiz b. Rajab, to accept the throne; and after

expressing a reluctance that may not have been totally assumed, he did so. The accession of Firuz Shah did

not go unchallenged. The claims of the the late monarch's nephew, Dawar Malik, were advanced by his

mother, Tughluq's daughter Khudawandzada, who was dissuaded by the amirs on the grounds of her son's

inexperience.97 In the capital the wazir Khwaja Jahan Ahmad b. Ayaz had set up as sultan an alleged child

of Muhammad's as Ghiyath al-Din Mahmud Shah.98 As Firuz Shah moved on Delhi, he was joined by a

great many notables who had deserted Khwaja Jahan. Eventually the wazir himself appeared in an attitude of humble submission. Firuz Shah was disposed to be merciful, but yielded to pressure from his amirs, who

were out for the old wazir's blood. Khwaja Jahan, despatched to his new iqta' of Samana, was overtaken

and executed by Shir Khan, its current muqta'. A few of his associates were likewise put to death;99 but the

fate of the child sultan he had enthroned is a mystery.

Regarding this affair the sources differ. The SJrat-i Firuz-Shahi, which refers to several later plots

against Firuz Shah in the vaguest of terms, is even less forthcoming about the reaction at Delhi to the news

of Muhammad's death, making no mention of the child sultan and merely condemning the treachery of the

wazir.100 The most plausible account is given by 'Afif. The wazir, who was now the sole member present in

Delhi of the triumvirate set up by Muhammad to head the government during his absence, heard reports not

simply of Muhammad's death but also of upheavals in which Firuz and Tatar Khan had disappeared. After

performing the mourning ceremonies both for the late sultan and for Firuz, to whom he was sincerely attached, the wazir enthroned a child of Muhammad and distributed largesse in order to buttress the infant

ruler's position. Only when it was too late to draw back did he learn that the troops in Sind had raised up

Firuz

97 She and her husband Khusraw Malik were later foiled in a bid to assassinate the sultan and were

punished: 'Afif, 45, 100-4. For the parentage of Dawar Malik and the confusion in the sources between him

and Khusraw Malik (who was actually his stepfather), see Hodivala, Studies, I, 309-10.

98 TMS, 120, is the only literary source to give the style of the infant monarch, for whose coins see

CMSD, 154 (nos 648-648B); J. G. Delmerick, 'Note on a new gold coin of Mahmud Shah bin Muhammad

Shah bin Tughluq Shah of Dihli', JASB, 43 (1874), 97-8. 99 TFS, 534-47. 'Afif, 57-78.

100

SFS, 12-13 (tr. Basu, JBORS, 22 [1936], 265ff.). For the plots, see ibid., 7-12 (tr. Basu, 101-7);

also TFS, 552, for an attempt to poison Firuz Shah.

~ Shah as sultan.101 'Afif appears to accept the boy as genuine;102 and he expressly challenges the

story that was current in his day - and retailed, for instance, by Barani - in which the wazir set up some

'bastard child' (walad al-zana'i) after learning of the accession of Firuz Shah, and scattered gifts with a

view to the imminent struggle for the throne.103 It is noteworthy that Barani is the sole author to claim that Muhammad had designated Firuz Shah as his heir (wali-'ahd).104

It is not easy to explain these discrepancies. The grounds for Barani's stance are especially

problematic. He is known to have suffered a loss of favour under the new sultan and been imprisoned for

some time in the stronghold of Bhatner.105 The earlier recension of his work is more out-spoken regarding

the dismissal and execution of Muhammad's servitors by Firuz Shah; the revised version, on the other hand,

strikes a more positive note, contrasting that ruler's leniency with the bloodshed that had been required to

ensure the triumph of previous Delhi Sultans.106 It therefore looks as if one of the purposes behind the

redrafting of the Ta'nkh-i Firuz-Shahi was to curry favour with the new monarch. 'Afif, writing well after

FIruz Shah's death, was perhaps under less pressure to lend legitimacy to his accession; although even he

retails stories in which saints as eminent as the sufi shaykhs Nizam al-Din Awliya' and 'Ala' al-Din

Ajudhani predicted Firuz Shah's sovereignty, and stresses that the caliphal patents that reached Firuz Shah came unsolicited, in contrast with the recognition that Muhammad had obtained only on request.107

Completing the revised version of his Ta'rikh-i Firuz-Shdhi in 758/1357, Barani was in a position

only to assess FIruz Shah's policies during his first few years. Yet he had little doubt what those policies

were. There had been no milder sovereign than FIruz Shah since the capture of Delhi; and no previous

sultan had avoided shedding blood to the extent that Flrtiz Shah had done with regard to Khwaja Jahan's

supporters;108 the harsh punishments of previous reigns were now discarded; spies and informers were a

101 'Afif, 50-3. TMS, 119-20, gives a similar but briefer version.

102 'Afif, 50, 60, 68, 396; cf. also 54, where the view of the army commanders in Sind, that

Muhammad had no son, is reported without comment.

103 TFS, 539; TFS1, Bodleian ms, fol. 212a, calls the child also ghulamzada. TMS, 120, describes

him as 'of unknown ancestry and non-existent lineage' (majhul al-nasab-u mafqiid al-hasab). al-Safadi,

Wafi, III, 172-3 (tr. Khan, 187), heard that Muhammad was incapable of fathering children. The boy is

accepted as a genuine son of Muhammad by Haig, 'Five questions', 365-72, and by Jamini Mohan Banerjee,

History of Firuz Shah Tughluq (Delhi, 1967), 15-16. Husain, Tughluq dynasty, 387-8, and B. P. Saksena, in

HN, 569-71, are noncommittal.

104 TFS, 532; cf. also 539.

105 Barani, Na't-i Muhammadi, cited in M. Habib, 'Life and thought of Ziyauddin Barani', in M.

Habib and A. U. S. Khan, The political theory of the Delhi Sultanate (Delhi, 1960), 162 (repr. in Nizami,

Politics and society, II, 348-9). See also TFS, 125, 554, 557, and other references in PL, I, 506. I. Habib,

'Barani's theory', 102.

106 TFS , Bodleian ms., fol. 217a; TFS, 547-52. Siddiqui, 'Fresh light', 78-9.

107 'Afif, 27-9, 273-4, 276.

108 TFS, 548, 551-2.

~ thing of the past.109 The soldiery enjoyed unprecedented ease: they were able to benefit from the revenues of their villages without even having to serve in the field.110 The new sultan's concern also for the

welfare of the 'religious class', to which Barani devotes a whole section of his work,111 had been

demonstrated at the very outset, in the course of his long journey from Thatta to Delhi. At Siwistan he had

restored to the 'ulama', the shaykhs and other notables the pensions, stipends and estates that Muhammad

had confiscated (presumably at the time of Qaysar's rebellion: below, p. 271) and bestowed alms on the

faqirs and wayfarers. At Uchch he rebuilt the dilapidated khanaqah of Shaykh Jamal al-Din and returned to

the shaykh's grandsons their estates and orchards which his predecessor had resumed to the khalisa. The

petitions of the people of Multan were granted, and gifts were made to the impoverished family of Shaykh

Farid al-Din at Ajudhan.112

In some measure, these can be viewed as the policies of a new monarch with an insecure title and

a consequent need to buy support. For this same reason - to avoid a recurrence of the troubles that had

afflicted his predecessor - the sultan made concessions to the nobility and the military class. It was

especially necessary for Firuz Shah to promote an image that contrasted with Muhammad's; and indeed the

policies he followed tell us a good deal about those of Muhammad which had aroused such resentment. In

his Futuhat the sultan himself reveals clearly the orthodox Islamic credentials for which he wished to be

remembered: the abandonment of draconian punishments; the abolition of uncanonical taxes; the

suppression of deviant forms of Islamic practice; the destruction of newly built Hindu temples; the

promotion of conversion to Islam among the Hindu populace; the foundation of new mosques and

madrasas; the repair of structures erected by past Muslim sovereigns; and humble attentiveness to Muslim

saints. Similar preoccupations - though with the addition of holy warfare against the infidel - are reflected

in the Sirat.113

Such attitudes might not necessarily have sprung from devotion alone. 'Afif's Ta'rikh-i Firuz-Shahi

reveals that it was Firuz Shah's practice to visit and pray at the shrines of saints and past sultans on the eve

of all his campaigns, as he did for instance before marching against Thatta;114 and his halts at shrines in

Sind had doubtless been designed to ensure victory over the faction of Khwaja Jahan. In any case we are

told later that the sultan made a point of visiting shrines whenever he was out riding.115 'Afif strongly

suggests, in fact, that FIruz Shah continued to identify himself with orthodox piety and with the interests of

the religious elite throughout his

109 Ibid., 557, 572-4.

110 Ibid., 553.

111 Ibid., 558-61.

112 Ibid., 537-9, 543.

113 For an interesting assessment of the sultan, see Khurram Qadir, 'Firoz Shah (Tughlaq): a

personality study', JCA 9 (1986), no. 2, 17-39.

114 'Afif, 194-6; and see also 230-1, 250.

115 Ibid., 371.

~ reign, even to the extent of having his head shaved like that of a sufi disciple (murshid) after the

death of his heir Fath Khan in 778/1376; it was immediately after this that he prohibited all practices in his

dominions that were contrary to the Shari' a.116

In military terms Firuz Shah's reign was undistinguished. He was obliged to acquiesce in the loss

of the Deccan and the far south, and his few campaigns to the east - against Bengal in 754/1353 and

760/1359 and Jajnagar in c. 761/1360 - achieved little. His sole successes were the subjection of the Hindu

ruler of Nagarkot (Kangra) in 766/1364-5 and the submission, after two invasions, of the Jams of Thatta

(767/1365-6). A whole section of 'Afif 's Ta'rikh is devoted to the sultan's abandonment of distant

campaigns (there were still forays to regions nearer at hand, such as the Sirmur hills, Katehr and Etawa)117 and his concentration on settling the affairs of state.118 But the reasons given vary. First 'Afif tells us that

when the wazir Khan Jahan (I) deflected him from invading the Deccan Firuz Shah promised not to lead an

army against his co-religionists again,119 a sentiment that recalls the inhibitions of Jalal al-Din Khalji.

Elsewhere 'Afif provides what seems like an alternative explanation for the abandonment of military

exploits: during the blockade of Thatta, Firuz Shah allegedly vowed that if he reduced the place he would

turn to other affairs.120 At yet later points in the biography, the sultan is said to have given up cam-paigning

after the death of the highly efficient and trusted Khan Jahan in 770/1368-9 (which would in fact have

occurred soon after the end of the Thatta enterprise).121 These various attempts to account for the sultan's

military inactivity in his later years suggest, in fact, that his biographer may have found the matter a source

of embarrassment. In the assertion that Firuz Shah's victories caused the people to forget war and to neglect weaponry, there is just a hint that his government undermined the Sultanate's military capacity.122

'Afif and the two authors who cover the entire reign, Sirhindi and Bihamadkhani, claim that the

era was characterized by prosperity, justice, clemency and security. Old men assured Sirhindi that there had

been no ruler more just, more merciful or more God-fearing since Nasir al-Din Mahmud Shah.123 The

cheapness and plenty of the reign, according to

116 Ibid., 372-3. The date is given as 777/1376: ibid., 379. But Fath Khan's death is later dated

Safar 778/June-July 1376, ibid., 493-4, although TMS, 131-2, supplies the date 12 Safar 776/23 July 1374.

117 Ibid., 134-5. 'Afif, 493, 497. The dates given are inconsistent.

118 Ibid., 261-7.

119 Ibid, 266.

120 Ibid, 216.

121 Ibid., 399, 424. For the year of Khan Jahan's death, which is variously given as 770 (ibid., 345)

and 772 (ibid., 422; TMS, 131), see Hodivala, Studies, I, 339, who opts for the earlier date.

122 ‘Afif, 23. See further the comments of Hardy, 'Force and violence', 178.

123 TMS, 140-1. 'Afif, 94, 99-100, 178-80, 193, 456, 512. Bihamadkhani, fol. 407b (tr. Zaki, 4).

See also the remarks in TFS, 553-4.

~ 'Afif, made Firuz Shah's subjects forget the prosperity even of 'Ala' al-Din's time; and whereas

'Ala' al-Din had brought about low prices by decree, under Firuz Shah they materialized without any effort

on the part of the government.124 Yet Bihamadkhani and Sirhindi wrote at a time when the Sultanate was a

mere shadow of its former self;' Afif, for his part, completed his biography in the wake of years of

internecine strife among Firuz Shah's descendants which had already erupted before he died, on 18

Ramadan 790/20 September 1388, and after the major cataclysm that was Temur's sack of Delhi in

801/1398.125 Thus Firuz Shah could be apostrophized, in terms evocative of the Prophet himself, as 'the seal

(khatm) of the sovereigns of Delhi'. More strikingly, perhaps, 'Afif presents the sultan as a holy man; and

the remark that the fall of Delhi ensued upon his death, with its hint that only his existence there had kept the city from destruction, forcefully echoes the idea of spiritual power (baraka), found in sufi literature,

that we noticed earlier.126 If Barani measured the opening years of the reign against the background of

Muhammad's regime, for these later authors Firuz Shah's day took on the colours of a golden age by

comparison with what followed.127

124 'Afif, 293-4.

125 Ibid., 133, where the sack is described as recent. But the fact that at 314-15 Temur is referred

to, not in opprobrious terms, but by the appellation 'Sahib-Qiran' ('Lord of the Fortunate Conjunction')

favoured in the Timurid sources, suggests that 'Afif wrote under the Sayyids (i.e. after 1414), who

acknowledged Timurid overlordship (below, pp. 318-19, 322). 126 'Afif, 21-2, 28. Digby, 'The sufi shaykh and the sultan', 77 and n.69. For an overview of cAfif's

treatment of the sultan, see Hardy, Historians, 41-51.

127 'Afif, 292-3. For the importance to 'Afif of Temiir's attack, see Hardy, Historians, 41 (and cf.

also 55).

~ CHAPTER 9

The KhaljT and Tughluqid nobility

The emergence of a new elite

We saw earlier (pp. 83-5) how at 'Ala' al-Din Khalji's accession the nobility was little changed

from that of the Ghiyathids. Only after a year or two did the new sultan move against the older aristocracy

which he had inherited from his uncle and a noble class emerge which differed substantially from that of

Balaban and Kayqubad. Barani divides 'Ala' al-Din's reign into three periods, of which the first was the era

of men who were closely linked with his seizure of the throne; the important figures of the second period

appear to have been largely bureaucrats; and the third, lasting for four or five years, was dominated by the

malign influence of the slave commander Kafur, by now the sultan's viceroy and hence generally called in

the sources 'Malik Na'ib'.1

The obscurity surrounding the origins of many of 'Ala' al-Din's nobles is perhaps only to be

expected. Some of the new elite would have been of Khalaj stock, and like his predecessor the new sultan

at first promoted close kinsmen, like his brother Almas Beg, now Ulugh Khan, who was made barbeg

(amir-hajib) and given the iqta' of Bhayana; subsequently, in 700/ 1301, he was granted the newly reduced

territory of Ranthanbor and Jhayin as his iqta'.2 Sanjar, entitled Alp Khan, who served 'Ala' al-Din as amir-i

majlis, was his wife's brother: at one point 'Isami says that 'Ala' al-Din had reared him since his childhood.

He held Multan for a time and was later transferred to the iqta' of Gujarat in c. 1310.3 Of 'Ala' al-Din's

brother's sons one, Sulayman Shah, became wakil-i dar and received the style of Ikit Khan, while another

was granted the title of Qutlugh Khan.4 A maternal nephew, Hizabr al-Din Yusuf, became Zafar Khan and

'arid.5

Apart from his kinsmen, the two principal amirs in the early years of the

1 TFS, 336-7.

2 Ibid., 242, 272, 283.

3 FS, 287, 288 (tr. 461, 463), for his iqta's; ibid., 338 (tr. 519), for his upbringing. The reading

HRBWN found alongside Alp Khan's name in TMS, 71, is an error for khusurpura ('father-in-law's son'), the

reading of one of the mss.; this term is also used of him in TFS, 242.

4 Ibid., 273. FS, 259, 279 (tr. 431, 453).

5 TFS, 240, 242, 248. TMS, 71.

171

~ reign were both men who had formed part of 'Ala' al-Din's entourage in Kara and Awadh prior

to his accession. 'Ala' al-Mulk, Barani's uncle, first acted as the new sultan's lieutenant in Kara and Awadh

and was then summoned to Delhi to become kotwal in succession to the former Malik al-Umara' Fakhr al-

Din.6 The other, Malik Nusrat Jalesari, obtained at 'Ala' al-Din's accession the title of Nusrat Khan; he may

well have been of relatively humble origin, as doubtless were many of these old associates.7 Nusrat Khan,

who was instrumental in securing enormous sums for the treasury from the elimination of the Jalali nobles,

was one of the new sultan's most trusted amirs, and it is significant that 'Ala' al-Din departed from the practice of his predecessors in making Nusrat Khan simulta-neously his na'ib and kotwal of Delhi. In the

following year he became wazir and surrendered the office of kotwal to 'Ala' al-Mulk, after which he

obtained the iqta' of Kara.8

It was a source of grim satisfaction to Barani that those of 'Ala' al-Din's henchmen who

participated in his uncle's murder all perished within a few years.9 Zafar Khan, who had played a

distinguished role against the invading Mongols, fell in battle with them in c. 1300.10 If we are to believe

the chronicler, who employs his uncle as a vehicle for advice to the sultan, 'Ala' al-Mulk was still alive at

the time of Qutlugh Qocha's attack; but he presumably died not long afterwards. Nusrat Khan perished

during the siege of Ranthanbor in 700/1300-1.11 To what further heights this powerful officer might have risen, had he survived, can only be guessed. Members of his family also attained prominence: a brother,

Malik 'Izz al-Din, amir-hajib to Ulugh Khan, had been killed by the neo-Muslim Mongols who mutinied on

the Gujarat expedition, and a nephew, Malik Fakhr al-DIn *Qochu, is subsequently found in possession of

the iqta' of Kara (probably in succession to his uncle) and in command of the troops of 'the east, Bengal and

Tirhut' in 702-3/1302-3, when he accompanied the dadbeg Fakhr al-Din 'A1i Jawna on the abortive

campaign against Arangal.12 But the subsequent history of this emerging aristocratic dynasty is unknown.

Certain of the sultan's kinsmen may well have proved a disappointment to him. BaranI heard that

Ulugh Khan died suddenly while planning an ambitious campaign to the far south: Tsami transmits a

rumour that he had been poisoned for reacting too swiftly to a false rumour of the sultan's

6 TFS, 250 (reading, with BL ms., fol. 130b, az mahlul-i malik al-umara-yi qadim), 257'. 7 Iqtidar Husain Siddiqi, The nobility under the Khalji Sultans', IC 37 (1963), 59-60.

8 TFS, 248, 249, 250, 272.

9 Ibid., 236-7. They included also Malik Asghari the sar-i dawdtdar and Malik Jawna the dadbeg,

of whom little is known, apart from their offices.

10 Ibid., 253-4, 260-1. FS, 259, 262-7 (tr. 431, 434-40).

11 TFS, 255-7, 266-72, for 'Ala' al-Mulk; 272 for Nusrat Khan. 12 'Izz al-Din: ibid., 252. Fakhr al-Din: TFS1, Digby Coll. ms., fol. 113a; TFS, 300 (reading QJW,

with BL ms., fol. 149a, for the JHJW of the text).

~ death.13 Ikit Khan aspired to emulate his uncle's success in seizing the throne. During a hunting

excursion at Tilpat on the march towards Ranthanbor (c. 1301), his men fired at 'Ala' al-Din, who was,

however, merely wounded. Duped by the sultan's paik guards into believing that 'Ala' al-Din was really

dead, Ikit Khan had himself proclaimed sovereign. But when 'Ala' al-Din appeared on the scene, the troops

rallied to him, and Ikit Khan was killed as he fled; his brother Qutlugh Khan was also put to death.14 Not

long afterwards, two of 'Ala' al-Din's sister's sons, 'Umar Khan and Mengu Khan, who held the iqta's of

Bada'un and Awadh respectively, were executed for treasonable designs.15

Our information regarding the amirs on whom 'Ala' al-Din relied during the central part of his

reign is relatively meagre; but the list given by Barani suggests that they belonged in the main to the

bureaucracy. They included Malik Hamld al-DIn, the son of 'Umdat al-Mulk Khwaja 'Ala-yi Dabir, who

became nd'ib-i wakil-i dar, and his brother Malik Izz al-Din, who was made chief secretary of the empire

(dabir-i mamalik): the brothers' rise seems to date from around the time of Ikit Khan's conspiracy and the

Ranthanbor campaign. Other major figures were Sharaf Qa'ini, the nd'ib-wazir, who is credited with

imposing a uniform system of tax assessment on an unprecedented number of provinces (see chapter 12);

'Ayn al-Mulk Multani, who had began his career as secretary (dablr) to Ulugh Khan; and Khwaja Nasir al-

Mulk Siraj al-DIn Hajji, the na'ib-i "ard-i mamalik, who later accompanied Kafur on his southern

campaigns.16 Although the ascendancy of these men may have stemmed from an increasing reluctance on

'Ala' al-Din's part to depend on his relatives, it is also clearly linked with his administrative and military reforms, which enabled him to maintain the armies that both repelled the formidable Mongol threat,

conquered a number of Hindu states in Rajasthan and the Yadava kingdom of Deogir, and plundered the far

south.

Appropriately enough for a sovereign whose reign was marked by numerous battles with invading

Mongols, 'Ala' al-Din's relations with Mongol amirs within India did not run smoothly. Barani suggests that

many or all of them forfeited their stipends.17 Certain of these 'neo-Muslim' commanders accompanied the

Gujarat expedition in 698-9/1299-1300 and mutinied when the sultan's generals tried to deprive them of

part of their plunder. The outbreak collapsed, and some fled to Karnadeva, the Vaghela

13 TFS, 283. FS, 281-2 (tr. 456-7).

14 TFS, 273-6. FS, 279-81 (tr. 453-5). A briefer account is given in IB, III, 185-6 (tr. Gibb, 641).

For a distorted account that reached Mongol Persia, see Qashani, 190-2.

15 TFS, 277-8.

16 For Barani's list, see ibid., 337. On Hamld al-DIn and his brother, see also ibid., 274-5, 282. For

Sharaf ,Qa'ini's activities, ibid., 288-9 (the correct reading QAYNY is found in BL ms., fols. 143, 167a); see

also Hodivala, Studies, I, 278. Khwaja Hajjl: KF, 82, 85, where his full style is given; RI, II, 56-60; TFS,

326, 328, 333.

17 Ibid., 334.

~ king of Gujarat, while others sought refuge at Ranthanbor. 'Ala' al-Din took a terrible vengeance

on their families in Delhi.18 In the following year, Ikit Khan in his bid for the throne drew support from

some neo-Muslim Mongol horsemen in his service; their subsequent fate is unknown.19 Still later, during

Kafur's Ma'bar campaign, a Mongol commander named Abachi planned to betray the Delhi forces to the

enemy and to kill Kafur. The plot failed, and the sultan had Abachi executed in Delhi. In reaction, the

Mongols in the capital, who allegedly numbered more than 10,000, conspired to kill 'Ala' al-Din and to

replace him with their own nominee, whereupon the sultan issued orders to his muqta's to arrest all the

Mongols in the empire and put them to death.20 The victims may have included 'Ali Beg and *Tartaq, who

had commanded the Mongol invading forces in 705/ 1305 (below, p. 227) and had been recruited into the sultan's service.21

It seems that Turkish slaves now played a more restricted role than under the Shamsid and

Ghiyathid monarchs. Only a few amirs - notably Ikhtiyar al-DIn Temur, who appears as muqta' of Chanderi

and Erach in an inscription of 711/1312 with the sobriquet 'Sultanf (i.e. a slave of the reigning sultan), and

Ikhtiyar al-DIn Tegin, muqta' of Awadh - are known from their names to have been Turks.22 The apparent

decline in the number of Turkish slave nobles may have been a matter of policy - a reluctance on 'Ala' al-

Din's part to allow Turkish ghulams the stranglehold on the administration that they had enjoyed in the

thirteenth century. It could also have been a reflection of the rising cost of such slaves, since Barani

complains that their price had risen prohibitively by his day;23 though this did not prevent the future sultan

Muhammad b. Tughluq from accumulating large numbers of Turkish slaves in the early 1320s (see below,

pp. 183-4).

The partial eclipse of Turkish slave amirs could well be connected with the rise of two new groups

of whom we first hear during this middle phase of 'Ala' al-Din's reign. Afghans had served Balaban and

Kayqubad (above, p. 62), and appear to have regularly formed part of the garrison troops of the Multan

province, where they are found both under Muhammad the 'Martyr Prince' and under Kushlu Khan in the

early Tughluqid era.24 But it is now that they first seem to have provided officers of high rank like Malik

Ikhtiyar al-Din *Mall, listed by BaranI among the sultan's nobles and later described as one of his great

maliks.25 The other category is 'Ala' al-Din's

18 Ibid., 252-3. FS, 253-5 (tr. 424-5).

19 TFS, 273.

20 FS, 296-7, 298-9 (tr. 470-1, 473-4). TFS, 334-6, setting the total number slain at 20 or 30,000.

TMS, 75, dates this episode in 697 and evidently confuses it with the mutiny on the Gujarat campaign.

21 KF,41.FS, 305 (tr. 481-2).

22 The readings in TFS (241, NKYN, JBAR; 323, BKTN), should be corrected from BL ms., fols.

125b-126a, 160b. For Temiir, see also Desai, 'Chanderi inscription'.

23 TFS, 314.

24 TS, IOL Persian ms. 412, fol. 52. IB, III, 322 (tr. Gibb, 712).

25 TFS, 240-1, 448. See I. H. Siddiqui, 'The Afghans and their emergence in India as ruling elite

during the Delhi Sultanate period', CAJ 26 (1982), 252 and n.45.

~ Indian slave officers. A later source alleges that 'Ala' al-Din possessed 50,000 slaves,26 of whom

the majority would have been Indians. The victorious campaigns by his forces against a number of major

independent Hindu kingdoms afforded greater opportunities for the acquisition of choice Indian slaves, and

it is in 'Ala' al-Din's reign that we first encounter their promotion to high office. The earliest to be

mentioned is Shahin, an obscure figure whom Isami calls the sultan's adopted son and Kafur's predecessor as na'ib. Put in command of Chitor on its capture in 703/1303, he later took fright following Ulugh Khan's

death and joined the exiled ruler of Gujarat.27 Malik Dinar, who served 'Ala' al-Din as shihna-yi pil,was

also an Indian slave.28 Malik Nanak, another slave, helped to save 'Ala' al-Din's life when his nephew Ikit

Khan made a bid for the throne in c. 1301, and was dkhurbeg and muqta' of Samana and Sunnam by

705/1305, when he defeated an invading Mongol army.29 Indian slave officers were not necessarily

converts to Islam: Amir Khusraw expressly refers to this engagement as the victory of an infidel over other

infidels.30

The most celebrated of 'Ala' al-Din's slave lieutenants, of course, is ' Kafur, an Indian captured

from his owner in Kanbhaya (Cambay) during the first invasion of Gujarat in 698/1299. Kafur, a eunuch,

acquired the nickname Hazardinari ('of the thousand dinars') from the price the sultan paid for him.31 His early career in 'Ala' al-Din's service is nowhere described, but he fought against the invading Mongols and

held the rank of bdrbeg by 706/1306-7, when he enjoyed the sultan's confidence sufficiently to be given

command of the army that reimposed tribute on the Yadava kingdom of Deoglr.32 Kafur's first known base

was Rapri, on the Yamuna, which was his iqta' by 709/1309-10;33 but towards the end of the reign he was

in command at Deoglr, which had by then been annexed to the Sultanate (p. 202 below); the date of his

appointment as na'ib is unknown.

During the final phase of the reign, Ala' al-Din was losing his grip: Barani regards as symptomatic

his appointment of Hamld al-Din Multani, a royal chamberlain and door-keeper (kalid-dar), as chief, qadi

of the empire.34 Barani's scattered observations elsewhere suggest that the sultan had ceased to trust the

majority of his higher servitors. He proved increas-

26 'Afif, 272.

21 FS, 281-3 (tr. 456-7).

28 TFS, 388-9. ■

29 Ibid., 273, 320, 323 (reading MANK, NAYK and TATK respectively, but cf. BL ms., fols. 142a,

158b; Hodivala, Studies, I, 243-4, 372, and II, 96, takes the name as 'Nayak'). KF, 38-9, confirming that he

was a personal slave (banda-yi khass). FS, 302-5 (tr. 479-81).

30 DR, 61. KF, 38. Siddiqi, 'Nobility under the Khaiji Sultans', 60 n.47. See also Amir * Khusraw,

Baqiyya Naqiyya, IOL Persian ms. 412, fols. 357b-358a.

31 TFS, 251-2. IB, III, 187 (tr. Gibb, 642), refers to him as al-Alfl, again in reference to the price

paid for him. That he was a eunuch (majbub) emerges from DR, 257, and TFS, 368. See generally S. Digby

'Kafur, Malik', Enc.Isl.2.

32 KF, 65; for other references to him as amlr-hajib, see ibid., 89, 114.

33

TFS, 328, 333. Yazdani, 'Inscriptions of the Khalji Sultans', 30.

34 TFS, 352; cf. also 298.

~ ingly unwilling to take advice, trying to supervise the conduct of all state business in person and

to this end, it appears, dispensing with the office of wazir, whose duties he fulfilled himself. Certain senior

officers, such as Malik Qiran the amir-i shikar and Malik Qirabeg, still enjoyed 'Ala' al-Din's favour; but

they had no power and were little more than courtiers. Experienced and skilled administrators were

removed, and in their place the sultan relied on those whom Barani terms lazy slaves (ghulambachagan)

and indiscreet eunuchs. The sultan also sought to. concentrate power in the hands of his own family and his

slaves: as a result, he promoted his pleasure-loving heir Khidr Khan prematurely and became too dependent

on Kafur.35 From the fact that the brothers Hamid al-Din and 'Izz al-Din were dismissed from office, and

Sharaf Qa'inl was put to death,36 it looks as if Kafur perceived these officers as a threat and prevailed upon

'Ala' al-Din to carry out a purge.

'Ala" al-Din's final months, already marred by illness, were clouded by a bitter rivalry between

Malik Na'ib Kafur and Alp Khan which in Barani's view destroyed his regime.37 As maternal uncle to the

sultan's heir Khidr Khan, Alp Khan had retained some power and influence almost to the very end of the

reign, since in what was clearly a bid to secure the succession the sultan married one of Alp Khan's

daughters to Khidr Khan and another to a younger son, Shadi Khan.38 But Alp Khan and his two sons-in-

law alike fell victim to the machinations of Kafur. The na'ib observed that the sultan was tiring of his chief

wife, Alp Khan's sister and Khidr Khan's mother, and set to work to undermine the influence of this family

group. 'Ala' al-Din was brought to sanction the murder of Alp Khan in the royal palace, and Khidr Khan

was first banished from court to Amroha and then imprisoned in Gwaliyor. The story that reached Persia

was that Khidr Khan, his mother and Alp Khan had poisoned 'Ala' al-Din, who was able, however, to execute them all before he died; and to some extent this is corroborated by Ibn Battuta, who heard that they

had conspired to replace the sultan with his son.39 The story may, of course, be nothing more than

propaganda circulated by Kafur.

The ailing sultan altered the succession in favour of a younger son, Shihab al-Din cUmar, whose

mother was the daughter of Ramadeva, the Yadava king of Deogir, and who was duly enthroned by Kafur

when 'Ala' al-Din died in 715/1316.40 It is tempting to see behind Kafur's coup d'etat an Indian faction,

comprising slaves fronted by a puppet sultan who was

35 Ibid., 334, 337-8, 367-8. TFS1 Digby Coll. ms., fol. 138a, says that Kafur was made wazir.

36 TFS, 337. 37 Ibid., 368. For what follows, see generally Lai, History of the Khaljis, 265ff. Apart from TFS,

368-9, the main sources are FS, 337-44 (tr. 517-24), and TMS, 79-81.

38 TFS, 368. FS, 336 (tr. 516).

39 Qashani, 193-4. IB, III, 187 (tr. Gibb, 641-2).

40 TFS, 374, reading RAMDYW for ZAYDH, as proposed by Hodivala, Studies, II, 106, and found in

BL ms., fol. 185b. The relationship is confirmed by FS, 343 (tr. 524).

~ himself half Indian; but the evidence does not permit us to do so. Kafur seems to have enjoyed

the cooperation of Kamal al-Din Gurg ('the wolf), whose family originated from Kabul.41 It was this officer

whom he sent to subjugate Gujarat when the province revolted on the news of Alp Khan's death; and

Sirhindi even has Kamal al-Din participating with Kafur in Alp Khan's murder.42

More probably, therefore,

the two groups we are dealing with represent merely the old 'establishment', perhaps centred on persons of

Khalaj origin, and a faction composed of relative newcomers, from widely differing backgrounds, which

looked to the na'ib for preferment.

Despite Kafur's activities, a sufficient number of 'Ala'I maliks survived to provide some kind of

continuity. They included not only Temur and Tegin but also two important figures associated with the middle period - Khwaja Hajji and 'Ayn al-Mulk Multani, who remained respectively 'arid and governor of

Malwa.43 Hushang, the son of Kamal al-DIn 'Gurg' who had been killed at the time of 'Ala' al-Din's death

while trying to put down a revolt in Gujarat, succeeded to his father's iqta' of Jalor.44 Qutb al-Din Mubarak

Shah gave Malik Dinar the style of Zafar Khan and sent him to govern Gujarat.45 It was also to the young

sultan's credit that he subse-quently appointed as governor of Gujarat the able and well-born Wahid al-Din

Qurayshi. Little is known regarding the background of the sultan's maternal kin, of whom Muhammad

*MtilaI became Shir Khan.46

Kafur's removal did not bring to an end the influence of Indian slave elements. Quite the contrary,

for during his brief reign of four years (716-20/1316-20) the new monarch came to rely inordinately on an

Indian slave Hasan, captured during the Malwa campaign of 705/1305. Hasan, "whom the sultan had

acquired from his na'ib-i khass-hajib, Malik Shadi, and who initially served as a member of the watch (pasban), obtained the dignity of wazir and the style of Khusraw Khan.47 Like Kafur, he aimed high, and

attempted to revolt while heading a campaign to the south; but when his colleagues reported his designs,

the infatuated sultan refused to believe them and had them punished.48 Khusraw Khan shortly murdered his

master (720/1320) and himself became sultan. What enabled him to do so was the fact that Qutb al-Din had

allowed him to accumulate a personal retinue of Parwari warriors from, his homeland in the region of

Bhilmal and Gujarat, whom he then introduced into the Hazar Sutun palace.49

41 See Desai, 'Jalor 'Idgah inscription', where he appears as Mahmud b. Muhammad b. 'Umar

Kabul!.

42 TFS, 369. TMS, 80. FS, 340-1 (tr.1520-2), has Kafur also sending Malik Dinar against the

Gujarat rebels.

43 NS, 100, 112. TFS, 379, 388.

44 Kamal al-Din: ibid., 369, 388. Hushang: ibid., 379-80 (with SWSMK in error for HWSNG);

Tughluq-Nama, 57, 65.

45 TFS, 381, 388-9; and cf. also 379. FS, 360 (tr. 558). 46 TFS, 381.

47 Ibid. TMS, 82-3, for Khusraw Khan aspasban; also 86. 48 TFS, 399-400.

49 Ibid., 402; and see Hodivala, Studies, I, 288. Khusraw Khan's brother Hasan had already

~ The Tughluqid coup and its beneficiaries

The origins of Tughluq, who had served 'Ala' al-Din for many years as muqta' of Deopalpur, are a

matter of controversy.50 No other source corroborates the assertion by certain near-contemporary authors

writing in the Mamluk empire that he was a slave.51 We can also, I suggest, discount the details obtained at

Lahore over three centuries later by Firishta, namely that Tughluq's father, also named Tughluq, was a

slave of Balaban who had married a Jat woman. Ibn Battuta, who arrived in Delhi within ten years of

Tughluq's death, learned that the late sultan was one of the Turks known as Qara'unas who inhabited the

territories between Sind and Turkestan (i.e. the Neguderis), and had reached India during the reign of 'Ala'

al-Din. 'Afif appears to have heard the same story some decades later, since he likewise describes Tughluq and his two brothers as coming from Khurasan in 'Ala" al-Din's time. But the earliest statement we have -

and the most deserving of credbnce - is that found in Amir Khusraw's Tughluq-Nama, composed in honour

of Tughluq's accession in 720/1320. Khusraw 'has Tughluq declare to the assembled grandees, following

the overthrow of Nasir al-DIn Khusraw Shah, that he was 'a nomad' (awara mardi) and had arrived in Jalal

al-pin Khalji's reign. This suggests that Tughluq was indeed of Mongol or Turco-Mongol stock, as Ibn

Battuta's informant claimed; he may have been a follower of the Mongol chief Alughu who entered Jalal al-

DIn's service in 691/1292 and settled near Delhi (above, p. 118).

The support for Tughluq's rising must have disappointed him. 'Isami, who claims that during his

march on Delhi he was joined by many 'Ala'i and Qutbi maliks,52 supplies no names. Apart from Bahram-i Ayba, muqta' of Uchch, whose father may have been one of Sultan 'Ala' al-Din's boon-companions

(nadiman),53 not a single governor is known to have rallied to his side. Of the amirs who were his

neighbours, Yaklakhi at Samana, an Indian, actually moved against Tughluq, but was repulsed and was

subse-quently killed by his own people. The same fate met Mughaltai, the amir of Multan, who had

declared for Khusraw Shah. Muhammad Shah Lur at Siwistan was forced by elements within the town to

support Tughluq, but in

■ attempted to gather Parwaris during a short visit to Gujarat as its governor: TFS, 396-7. TFS1,

Bodleian ms., fol. 173b/ Digby Coll. ms., fol. 147b, suggests that the Parwaris came from Jalor as well as

Gujarat. See S. R. Sharma, 'Nasir-ud-din Khusru Shah', in Mahamahopddhyaya: Professor D. V. Potdar ...

commemoration volume (Poona, 1950), 70-81, for this warrior group.

50 For what follows, see R. C. Jauhri, 'Ghiyathu'd-Din Tughluq - his original name and descent', in

Horst Kriiger (ed.), Kunwar Mohammad Ashraf: an Indian scholar and revolu-tionary 1905-1962 (Berlin,

1966), 62-6.

51 Ibn Abi'l-Fada'il, ed. Kortantamer, 27 (German tr. 104). al-Safadi, Waft, III, 172 (tr. Khan, 187).

52 FS, 381 (tr. 584).

53 Rukn al-DIn Ayba: TFS, 358 (reading DABYR to be corrected from BL ms., fol. 177b: 'YBH).

That Ayba was the name of Bahrain's father is clear from FS, 388 (tr. 593).

~ the event arrived too late to help him. Further afield, Malik Hushang at Jalor was unwilling to

commit himself, and 'Ayn al-Mulk Multani joined Khusraw Shah only to desert him on the eve of battle

and retire to his iqta's of Dhar and Ujjain in Malwa.54 Tughluq's adherents were kinsmen like his son Malik

Jawna (the future Sultan Muhammad), a son-in-law Malik ShadI, and two nephews, Asad al-Din Arslan

and Baha' al-Din Garshasp; or subordinates like Yusuf, Tughluq's na'ib at Deopalpur, and 'Ali-yi Haydar.55

Otherwise his army was made up of outsiders. Amir Khusraw's characterization of them - 'mostly from the

Upper Country (iqlim-i bala), neither Indian nor Indian chiefs (Hindu-wala): Ghuzz, Turks, and Mongols

of Rum and Rus ... Tajiks from Khurasan of pure stock'56 - is more than a trifle disingenuous. It ignores the

Khokhars under their chiefs *Samaj Rai and Gul Chand, to whom 'Isami attributes much of the credit for

the victory at Sarsati,57 but whose presence on Tughluq's side was difficult to reconcile with the rhetoric of

Holy War. Tughluq's affinity, in other words, was markedly regional; his lieutenants were commanders

who had fought alongside him on the Mongol frontier, sometimes themselves Mongol renegades, or Hindu warlords who were his close neighbours in the western Panjab.

Tughluq's following is decidedly less impressive than that of his antagonist. The nucleus, of

course, comprised Parwaris, headed by Khusraw Shah's maternal uncle *Randhaval. But among the

commanders whom Khusraw Shah sent to check the rebels at Sarsatl were Temur, the muqta' of Chanderi,

Qutlugh the amir-i shikar, and *Tulabugha Bughda.58 When Tughluq pushed through to Delhi, Khusraw

Shah met him at the head of forces that included *Tulabugha Bughda, *Tulabugha Nagawri, Tegin the

muqta' of Awadh, Ikhtiyar al-DIn Sunbul the amir-hajib, Kafur the keeper of the seal (muhrdar), and Qabul

the supervisor of the market (shihna-yi manda).59 None of these nobles was an upstart promoted by the

usurper: all had held office under Qutb al-DIn Mubarak Shah, and several - Temur, Tegin, Qutlugh and

Qabul - had served his father 'Ala' al-Din before him. Support of this calibre belies the traditional view of Khusraw Shah as a widely hated infidel and of his rival as the avenger of the Khalji dynasty and the saviour

of Islam.

Conscious, perhaps, of the relatively narrow support he had enjoyed, Ghiyath al-Din Tughluq after

his accession took care to draw in 'Ala'i

54 See Husain, Tughluq dynasty, 38-41. Muhammad Shah Lur appears among the nobles of 'Ala'

al-Din's reign: TFS, 24Q (omitting 'Lur', but cf. BL ms., fol. 125b). For Yaklakhfs Indian extraction, see

Tughluq-Nama, 68.

55 Ibid., 95. FS, 380, 382, 383 (tr. 582, 585, 586).

56 Tughluq-Nama, 84.

57 FS, 378, 379-80, 382, 384, 385 (tr. 580, 582-3, 585, 588-9). For the SHJ of the printed edition,

IOL Persian ms. 3089, fol. 208b, has SMJRAY.

58 Tughluq-Nama, 101-2. FS, 379-80 (tr. 582-3).

59 Tughluq-Nama, 118 (reading BKYN in error for TGYN). FS, 382, 383, 386 (tr. 585, 586, 589-90).

TFS, 420.

~ maliks and to secure their good will with offices and iqta's.60 Khwaja Hajji was retained as

'arid,61 and 'Ayn al-Mulk Multani remained governor of Malwa, though neither seems to have survived

Tughluq (this 'Ayn al-Mulk is to be distinguished from Ibn Mahru, who later bore the same title: see

appendix IV). But the alliance with many of the sultan's erstwhile colleagues seems to have been an uneasy

one. Tensions emerged in 721/1321-2, when Tughluq deputed a number of amirs from the old regime,

along with those of his own creation, to accompany his son and heir, now styled Ulugh Khan, to Arangal.

As the result of an intrigue which involved principally a poet in Ulugh Khan's service, named 'Ubayd, and

which the sources do little to elucidate, Temur, Tegin and Kafur (who had exchanged his office of muhrdar

for that of wakil-i dar) were easily brought to believe that Ulugh Khan planned to do away with them, and

deserted with their contingents, thus jeopardizing the entire campaign. Ulugh Khan extricated himself, and

troops were sent against the disaffected amirs. Temiir and Tegin were both killed while seeking refuge in Hindu territory; Kafur was taken prisoner and executed in Delhi; their families were all put to death.62 All

the amirs in question had served 'Ala" al-Din; it may be significant that even 'Ubayd had done so if, as is

likely, we can identify him with the 'Ubayd-i Hakim mentioned by Barani among that sultan's boon-

companions.63 More importantly, some of these amirs, as we noticed earlier, had supported Khusraw Shah

in 720/1320. It is hard to resist the suspicion that the episode afforded the sultan a convenient pretext for

eliminating powerful noble households in which he felt unable to repose complete trust. With their removal

a significant number of the leading figures of 'Ala' al-Din's era left the stage. ,

As might be expected of a monarch who had come to power with the aid of elements from the

north-west, Tughluq favoured officers from those parts. Burhan al-Din, who obtained the post of kotwal

and the style of 'Alim Malik, was the founder of an important noble family which had settled at Hansi but

originated from Ghazna.64 Of his sons, Kamal al-Din became chief qadi and Sadr-i Jahan under Muhammad, and his high standing in the empire is attested both by Ibn Battuta and by al-'Umari's

60 Ibid., 426. TMS, 92.

61 For Khwaja Hajji, see TFS, 438; FS, 395 (with CACY in error).

62 Prasad, Qaraunah Turks, 29-33, and Husain, Tughluq dynasty, 65-9, summarize the data in the

various sources. TFS, 448, 449, lists among the deserters the Afghan malik Ikhtiyar al-Din Mall (printed

text has MX, but cf. TFS1, Bodleian ms., fol. 183b/Digby Coll. ms., fol. 155a, with ML); but FS, 395-6 (tr.

603), says that he remained loyal. For Kafur's rank, see ibid., 394, 400 (tr. 599, 606).

63 TFS, 360.

64 Ibid., 424, for Burhan al-DIn. For the family's origins, see IB, III, 143, 161 (tr. Gibb, 617, 628).

Burhan al-DIn's wife, the mother of Kamal al-DIn, was the sister of Mawlana Fakhr al-DIn Hansawi: Siyar,

274.

~ informants.65 Another was Qiwam al-Din, who served as na'ib-wazir at Deogir and then, under

Muhammad b. Tughluq, was entitled Qutlugh Khan and promoted to wakil-i dar. At the time of the sultan's

abortive Ma'bar campaign in c. 1335, he was once more sent to Deogir (by then renamed Dawlatabad),

where he remained in authority for ten years. His recall in 745/1344-5 appears greatly to have undermined Muhammad's authority in the Deccan province and contributed to its secession three years later.66 Qutlugh

Khan's son Muhammad had received from his name-sake the titles of Alp Khan and Nizam al-Mulk,

together with the iqta' of Gujarat. He did not, apparently, hold this position for long, and in the late 1330s is

found deputizing for his father in Dawlatabad during the operations against the rebel Nusrat Khan.67 A third

son of Burhan al-DIn, Nizam al-Din, appears as one of Tughluq's maliks and subsequently, under

Muhammad, as 'Alim al-Mulk and governor of Bharuch (Broach).68 From there he was transferred to

Dawlatabad to replace his brother Qutlugh Khan temporarily, but was taken prisoner by rebels and later

released and sent to Delhi.69

The clan Abu Rija, another new lineage which seems to have attained prominence under the

Tughluqids, is again expressly said to have originated from the 'upper country' (mulk-i bala), i.e. the north-

west. One of its members, Mujir al-Din, had already been made nd'ib-wazir at Deogir by Qutb al-Din Mubarak Shah Khaljl.70 He was still to be found in the region at the time of the Arangal campaign, when he

furnished Ulugh Khan with valuable aid in overthrowing the mutinous nobles. This may have earned him

the future sultan's trust, for Ibn Battuta refers to him as one of 'the great amirs' of Muhammad b. Tughluq's

reign and Barani lists him among the evil influences on that monarch. He served Muhammad loyally,

participating with his forces in the campaigns against the rebel Baha' al-Din Garshasp and against 'Ayn al-

Mulk Ibn Mahru ten years or so later, when he was governor of Bhayana.71 Husam al-Din Abu Rija and

Shihab al-Din Abu Rija were probably his brothers. The latter obtained the rank of 'king of the merchants'

(malik al-tujjar) and the iqta' of Nawsari on Muhammad's

65 TFS, 454. TMS, 98. IB, III, 161, 215, 229 (tr. Gibb, 628, 657, 664). MA, ed. Spies, 16 (German

tr. 41)/ed. Fariq, 30 (tr. Siddiqi and Ahmad, 41). 66 TMS, 98. TFS, 454, 481, 501-2. FS, All, 426 (tr. 648, 653), shows that he had remained at

Deogir in the early years of Muhammad's reign.

67 TMS, 98. FS, 477l (tr. 717). He does not appear among Muhammad's maliks in the printed text

of TFS, but cf. BL ms., fol. 225b.

68 TMS, 98. TFS, 502. FS, 495, 503 (tr. 739, 749).

69 Ibid., 519 (tr. 770). TMS, 111.

70 FS, 369 (tr. 569). TFS, 398, elucidated by Hodivala, Studies, I, 287-8; in the list of Qutb al-Din's

maliks in TFS, 379, he appears as 'Fakhr al-DIn', but cf. BL ms., fol. 188b. See 'Afif, 454, for the

provenance of this family.

71 TFS, All. FS, 397-9, 427, 473 (tr. 603-5, 653, 711). TMS, 101. IB, III, 230, 318, and IV, 5 (tr.

Gibb, 665, 710, 775), commenting also on his cruelty. For a brief biography, see Hodivala, Studies, I, 287.

~ accession.72 Husam al-Din was mustawfi under Ghiyath al-Din Tughluq and was detailed to

secure Malik Tegin's household in the aftermath of the mutiny at Arangal. He retained his office into the

reign of Muhammad, who conferred on him the style of Nizam al-Mulk: sent to Lakhnawti as na'ib-wazir,

he helped to put down the first revolt of Fakhr al-Din ('Fakhra') at Sunarga'un (c. 1335-6).73 A nephew of

Mujlr al-Din, Shams al-Din, who was to acquire notoriety by his activities as mustawfi under Firuz Shah, was Husam al-Din's son.74 Another member of the family corresponded with Ibn Mahru, and some of the

Abu Rija clan went on to serve the independent Gujarat sultans in the early fifteenth century.75

The aristocracy under Muhammad b. Tughluq

Little is known of the fate of Tughluq's sons, none of whom appears to have survived

Muhammad's reign. One, Mubarak Khan, acted in a judicial capacity during the new reign; but Ibn Battuta

heard that another, Mas'ud Khan, was put to death, perhaps because his mother was a daughter of 'Ala' al-

Din Khalji.76

Firuz (the future sultan), the son of the late monarch's brother Rajab, served Muhammad as

barbeg.11 Two adopted sons of Tughluq certainly enjoyed considerable favour: Tatar Malik (actually the son of a Mongol prince who had invaded India during Tughluq's lieute-nancy at Deopalpur) attained some

prominence, despite temporary banishment following a quarrel with the sultan, while Bahram Khan was

entrusted with the government of Sunarga'un.78

Virtually half of the appointments made by Muhammad on his enthronement went to men who are

known to have originated from the north-west, and included the amirs he had inherited from his father and

who had played a leading role in Tughluq's revolt against Khusraw Shah.79 But within a few years

Muhammad was confronted with insurrections by two of these men, Baha' al-Din Garshasp and Kushlu

Khan, which seem to have been provoked by his attempts to intensify his authority in the provinces

72 TMS, 98.

73 TFS, 455 (cf. BL ms., fol. 226a). TMS, 94, 98, 104.

74 'Afif, 451, 454. Anonymous, Ghunyat al-Munya, ed. Shahab Sarmadee (Delhi, 1978), 6-7, gives

his full name, Shams al-Dawla wa'l-Din Ibrahim-i Hasan: Husam al-Din is known to have been called

Hasan.

75 IM, 157-9 (Malik Baha' al-Din Nasr-Allah, na’ib-i khass-hajib). Z. A. Desai, 'Inscriptions of the

sultans of Gujarat from Saurashtra', EIAPS (1953-4), 51.

76 Mas’ud Khan: IB, III, 292 (tr. Gibb, 696). Mubarak Khan: ibid., Ill, 230, 287-8 (tr. 664, 694);

mentioned in TFS, 454. 77 Ibid. 'Afif, 42. TMS, 98.

78 Biography of Tatar Malik in 'Afif, 388-94. He had held the iqta' of Zafarabad under Tughluq

(TFS, 428, 451), and was subsequently styled Tatar Khan by Muhammad's successor (TMS, 124). Later the

author of various legal works, he spoke fluent Arabic: IB, III, 281 (tr. Gibb, 690). On Bahram Khan, see

FS, 422, 444, 472 (tr. 648, 673, 709); TFS, 480; IB, III, 230, 317 (tr. Gibb, 665, 709), wrongly calling him

Muhammad's brother's son (Gibb's tr., 665 n.36, confuses Bahram Khan with Tatar Khan).

79 This emerges from the list given in TMS, 98.

~ (pp. 256-7 below). This may have provided the impetus to recruit a new body of servitors. He seems to have reposed great confidence in Ahmad b. Ayaz, whom he made wazir in 732/1331-2 with the

style of Khwaja Jahan and who is found on a number of occasions leading military forces against rebels.80

Khwaja Jahan served Muhammad loyally for the next twenty years, only to fall from power by becoming

the focus of resistance to Firuz Shah's accession. When Muhammad left Delhi for the last time, he

delegated authority in the capital to Khwaja Jahan, his cousin Firuz and Malik Qabul 'Khalifati' (also

known as 'Malik Kabir').81

As undisputed heir-apparent throughout his father's reign, Muhammad seems to have built up a

power-base of his own. Nigam's assertion that the slave system did not receive much encouragement during

Muhammad's reign82 is simply at variance with the testimony of our sources. Among his most trusted amirs

was Malik Qabul, his slave and probably an Indian; and we know that the sultan also recruited black slaves (Habashis), one of whom, presumably, was Badr al-Habashi, his governor at 'Alapur.83 Turks may now

have attained some prominence once more. Ibn Battuta heard that Muhammad had alarmed his father by

amassing a body of Turkish mamluks. Since the Moroccan traveller found 4,000 of them stationed at

Amroha alone, the total figure of 20,000 for Muhammad's Turkish slaves transmitted by al-'Umari is

probably too low.84 Ibn Battuta's vivid description of Muhammad's processions suggests that many of his

amirs may have been mamluks.85 We know the names of only a few of these Turks who rose to high office.

'Imad al-Mulk Sartiz, who became 'arid and governor of Multan and was later transferred to the Deccan,

where he fell fighting against the dmiran-i sada, was a slave and probably a Turk.86 To judge from

80 Date of his appointment: Siyar, 218. As a military commander: TFS, 481; FS, 425-31, 471 (tr.

651-8, 707-8); IB, III, 318, 324,;332-3, 348-9 (tr. Gibb, 710, 713, 716-17, 723-4). For his ancestry, see p.

189 below.

81 TFS, 509, 522. 'Afif, 50, 452.

82 Nigam, Nobility, 85.

83 Qabul: TFS, 493; IB, I, 365, and III, 230 (tr. Gibb, 226, 665), and passim. Habashis: ibid., IV,

31 (tr. Gibb and Beckingham, 786); later, IV, 59-60 (tr. 800), he refers to a guard of fifty Habashi men-at-

arms who embarked with him at Gandhar.

84 Ibid., Ill, 211 (tr. Gibb, 654); and see III, 439 (tr. Gibb, 763), for Amroha. MA, ed. Spies, 13

(German tr. 38)/ed. Fariq, 25 (tr. Siddiqi and Ahmad, 37); for another reference, to 12,000 mamluks

accompanying the sultan, ibid., ed. Spies, 19 (tr. 45)/ed. Fariq, 37 (tr. Siddiqi and Ahmad, 44). IB, III, 334

(tr. Gibb, 717), says that mamluks accompanied Muhammad on his Ma'bar campaign in c. 1335. But we

cannot be certain that these were Turks, since the same author employs the term for slaves whom we know

from other sources to have been Indians: see, e.g., ibid., Ill, 190, 191 (tr. Gibb, 643).

85 Ibid., Ill, 231 (tr. Gibb, 665).

86 Ibid., Ill, 44, 94 (tr. Gibb, 562-3, 593): at the latter place Gibb read 'arid-i mamalik, 'inspector-

general of the mamluks', for 'drid-i mamalik; but see the French editors' note at IB, III, 458-9. Although

described both here and at III, 107-9 (tr. Gibb, 600), as governor of Sind, he was not appointed to this post until after Shahu Lodi's revolt in c. 1337: see TMS, 107 (with the impossible year 744). Ibn Battuta may

thus have confused two visits to the province: see C. F. Beckingham, 'Ibn Battuta in Sind', in Khuhro (ed.),

Sind through the centuries, 140-1. Firishta, I, 522, calls Sartiz a 'Turkmen'.

~ his name, Malik Qiran Safdar Malik Sultani certainly was.87 Another Turk, lastly, was Taghai,

who passed from Malik Qiran into the possession of Sultan Muhammad and was promoted to be shihna-yi

bargah;88 his revolt in Gujarat towards the end of the reign proved the most intractable that the sultan had

to face.

At this time the Sultanate still served as a magnet for dispossessed princes, adventurers and

opportunists from the west. Kushlu Khan's rebellion in Sind had drawn on 'Turks, Afghans and the men of

Khurasan'.89 Al-'Umari speaks of Turks, natives of 'Khita' (literally 'northern China', but doubtless Mongolia) and Persians in the sultan's own army, and Ibn Battuta refers more than once to the 'amirs of

Khurasan' (see below, p. 263) among Muhammad's officers.90 What particularly attracted great numbers of

immigrant notables was Muhammad's proverbial munificence;91 the story of his generosity to Sayyid 'Adud

al-Din of Yazd, an envoy from Mongol Persia, for example, gained wide currency.92 Ibn Battuta, himself a

beneficiary of this policy, describes how foreigners were promoted to governorships and to high office, and

were treated with the greatest distinction, being addressed, on Muhammad's express instructions, by the

special title of 'aziz ('honourable one').93 In c. 733/1332-3, Nizam al-Din, a scion of the former ruling

dynasty of Qays in the Persian Gulf, arrived at Muhammad's court, where he spent two years in a vain

effort to secure the sultan's aid in recovering his patrimony.94 A few years later, Ibn Battuta found Hajji

Ke'iin, a brother of the Ilkhan Musa, as the sultan's guest: he returned to south-western Persia in 743/1342

and was killed while endea-vouring to occupy Shabankara.95 Muhammad is known to have sent agents 87 See Hodivala, Studies, I, 300-1. For Tu. qiran, 'one who slaughters', see above, p. 63, n.16.

88

The fullest account of his career is to be found in SFS, 19-28 (tr. Basu, JBORS 23 [1937], 97-

106). For his office, see IB, III, 235 (tr. Gibb, 667). His name is Tu. taghai, 'maternal uncle': Clauson,

Etymological dictionary, 474.

89 IB, III, 322 (tr. Gibb, 712).

90 MA, ed. Spies, 13 (German tr. 38)/ed. Fariq, 24 (tr. Siddiqi and Ahmad, 37). IB, III, 344, 348 (tr.

Gibb, 721, 723); cf. also III, 332 (tr. 716), for 'Khurasanis'.

91 Ibid., II, 72-7, and III, 97-9, 243-66 passim, 270, 279, 284 (tr. Gibb, 311-13, 595-6, 671-83

passim, 685, 689, 692). MA, ed. Spies, 22-5, 38 (German tr. 48-51, 65-6)/ed. Fariq, 41-6, 70 (tr. Siddiqi and

Ahmad, 48-50, omitting, 48, a sentence about foreigners; 63-4).

92 Shabankara'i, 288-9. MA, ed. Spies, 22 (German tr. 48-9)/ed. Fariq, 42-3 (tr. Siddiqi and

Ahmad, 47-8). al-Safadi, Wafi, Biblioteca dell'Accademia dei Lincei, Rome, ms. Fondo Caetani 21, 435-6.

A brief allusion in TFS, 461.

93 IB, III, 97-8, 243 (tr. Gibb, 595, 671).

94 Jean Aubin, 'Les princes d'Ormuz du Xllf au XVe siecle', JA 241 (1953), 105; Shabankara'i, 219.

This was the dynasty to which Siraj-i Taqi belonged (below, p. 208).

95 IB, III, 256-8 (tr. Gibb, 677-9). Jean Aubin, 'La question de Sirgan au XIIIe siecle', Studia

Iranica 6 (1977), 289, citing a passage found only in the Tabriz ms. of the third redaction of ShabankaraTs

Majma', which was used by the anonymous author of a general chronicle in Leiden University ms. Or.

1612, fol. 357a, and by Natanzi, Muntakhab al-Tawarikh (816/ 1413-14), partial edn by J. Aubin, Extraits

du Muntakhab al-tawarikh-i Mu'ini (Tehran, 1957), 9-10.

~ to the Persian Gulf to recruit Arab amirs and their followers into his service.96 In part this lavish

patronage was linked to his expansionist designs in what is now Afghanistan (see chapter 13), and the many notables from Mongol territory who arrived with Ibn Battuta included Khudawand-zada Qiwam al-

Din, qadi of Tirmid, his cousin Ghiyath al-Din, two grandees from Transoxiana, and Bahram, malik of

Ghazna; later there arrived two Mongol amirs, *Qabtagha and Ahmad-i Iqbal, of whom the first was

reputedly descended from the Mongol commander Temur who had overthrown Balaban's son in

683/1285.97

Despite Ibn Battuta's testimony that Muhammad preferred foreigners to the indigenous

aristocracy, and that the 'Indians' in turn hated the immi-grant 'Khurasani' nobles,98 there is clear evidence

that the position was more complex; the sultan's favour extended to a much wider clientela and native

Indians, like the future rebel 'Ayn al-Mulk Ibn Mahru (see appendix IV), also benefited from his generosity

and trust.' The Moroccan himself was on friendly terms with the muhrdar 'Abu Muslim', one of the many

sons of the rai of Kampila whom Muhammad had maintained at his court since the conquest of that territory.99 More notable was *Kannu, a Brahman taken prisoner to Delhi on the conquest of Tilang in c.

1322, who entered his service and embraced Islam, receiving the name Maqbul and subsequently the style

of Qiwam al-Mulk. Appointed governor of Multan by Muhammad on the suppression of Kushlu Khan's

rising in 728/1327-8, he briefly governed Tilang until its revolt in c. 1336, and later became deputy to the

wazir Khwaja Jahan Ahmad b. Ayaz. Following Firuz Shah's accession, he obtained the style of Khan

Jahan and succeeded Khwaja Jahan as wazir, an office he retained until his death and transmitted to his

son.100

In general, however, Barani stigmatizes Muhammad's Indian servitors as Jowborn.101 They

included the notorious 'Aziz Khammar ('the Vintner'), on whom the sultan conferred the government of

Malwa.102 A number of them were non-Muslims and - for all the chronicler's jaundiced remarks else-where about drapers (bazzazan), goldsmiths and the like103 - were probably

96 IB, IV, 104 (tr. Gibb and Beckingham, 818-19).

97 Notables from Transoxiana: ibid., Ill, 374-5, 394-5 (tr. Gibb, 735, 743-4). Bahram: ibid., Ill,

264-5 (tr. Gibb, 682). Ahmad and *Qabtagha: TFS, 520, 584-5; for the second name, probably Mo.

qabtagha, 'purse', 'bag', see Lessing, Mongolian-English Dictionary, 899. P. Jackson, 'The Mongols and the

Delhi Sultanate in the reign of Muhammad Tughluq (1325-1351)', CAJ 19 (1975), 147-8.

98 IB, III, 344, 349 (tr. Gibb, 721-2, 724). 99 Ibid., Ill, 320-1 (tr. Gibb, 711).

100 Multan: TMS, 98, 101. Tilang: TFS, 481 (reading QBWL to be amended as in BL ms., fol. 238a),

484. Naib-wazir. ibid., 454, 512. Biography in 'Afif, 394-424. He is to be distinguished from two of

Muhammad's other maliks, Qabul (p. 183 above) and Muqbil: see Hodivala, Studies, II, 112, 115-16.

101 List in TFS, 504-5 (to be corrected from BL ms., fol. 249b). FJ, 167-8.

102 TFS, 501-2, 503, 504.

103 FJ, 180-1; see also 295-302 (at 298 the reader is warned not to be taken in by their

administrative skills).

~ kayasthas, members of an administrative class, like Ratan, described as ‘a person skilled in

calculation and writing', who was entrusted with the fiscal administration of Sind; Bhiran, auditor

(mutasarrif) at Gulbarga; Samara Singh, who became governor of Tilang; and Dhara, whom Muhammad

sent to Dawlatabad as deputy wazir in 745/1344-5, just a matter of months before his authority there

disintegrated and the province seceded under the Bahmanid dynasty.!04

The era of Firuz Shah

In his first recension Barani speaks of the number of maliks from the previous regime who were brought low at Firuz Shah's accession.105 In fact, however, Maqbul was simply the most highly favoured

among a significant number of men who came over to the new monarch from Khwaja Jahan during the

early weeks and were retained in positions of trust. Husam al-Din, son of Malik Nuwa, became na'ib of

Awadh and received the style of Husam al-Mulk.106 Malik Mubarak, the son of Muhammad's leading amir

Malik Qabul Khalifati, served as silahdar-i khass and later wakil-i dar, surviving Firuz Shah himself.107

Even A'zam Malik Shaykhzada Bistami, who as one of the associates of Khwaja Jahan had been banished

from Firuz Shah's territories, was later pardoned when he reappeared with a caliphal robe, and was restored

to favour with the style of A'zam Khan.108 And although BaranI mentions - with ill-disguised relish - how

the new sultan dismissed the foreigners who had flocked to Muhammad's court from Herat, Sistan, Aden

and Qusdar in expectation of rewards,109 Firuz Shah's nobles included also some of the most distinguished

immigrants of the previous reign. Khudawandzada Qiwam al-Din Tirmidi, Muhammad's na'ib-i wakil-i

dar, became Khudawand Khan and wakil-i dar, while his nephew was entitled Sayf al-Mulk and made amir-i shikar-i maymana.110 The Mongol amirs *Qabtagha and Ahmad-i Iqbal, too, enjoyed Firuz Shah's

favour, and Ahmad's son Husayn in turn served the sultan and married his daughter.111

104 Ratan: IB, III, 105-6 (tr. Gibb, 599). Bhiran: FS, 485 (tr. 726-7). Samara Singh: K. H.

Kamdar, in Proceedings of the 7 th all India oriental conference (Baroda, 1933) (Baroda, 1935), 629-33.

Dhara: TFS, 501. Others are named ibid., 504-5. On this class, see generally Yusuf Husain, 'Les Kayasthas

ou "scribes", caste hindoue iranisee, et la culture musulmane dans l'Inde', REI\ (1927), 455-8.

105 TFS 1 Bodleian ms, fol. 217a. Siddiqui, 'Fresh light', 78.

106 TFS,528 (to be completed by the still slightly corrupt reading in BL ms., fol. 261a). TMS, 133. Bihamadkhani, fol. 417a (tr. Zaki, 22).

107 Ibid., fols. 416b-417a (tr. Zaki, 21). cAfif, 287, 338, calling him ambiguously Mubarak-i

Kablr.

108 TMS, 127-8; cf. also cAfif, 281; and for his partisanship of Khwaja Jahan, TMS, 120, 123, and

TFS, 543, 545.

109 TFS, 538. 110 Ibid., 454, 580. TMS, 124.

111 TFS, 527, 544, 584-5. cAfif, 280. TMS, 140, for the marriage.

~Like his predecessors Firuz Shah built up a corps of amirs of his own creation. One of the most

significant long-term developments of the reign was the accumulation of offices and iqta's in the hands of

his slaves. They are referred to by different authors, under the events of the following reigns, as 'Turkish'

slaves and' amirs '(bandagan-i turk, umara-yi atrak) and as 'Hindustanis'. The apparent contradiction may

be resolved if it is assumed that many of them were of eastern Indian provenance: thus Juziani, over a

century earlier, had written of the natives of Tibet and Arakan as 'Turks'.112 According to cAfif, Firtiz Shah

made greater efforts to acquire slaves than any of his predecessors: provincial governors were under orders

to forward the choicest slaves to court as part of their annual gift to the sultan, and the total number of royal

slaves rose to 180,000. Of these, 40,000 were in attendance at court or formed part of Firuz Shah's retinue;

the remainder were ocpupied in a variety of tasks, some of them being taught a skilled craft.113 The royal ghulams became such an important element in the state that responsibility for their affairs was transferred

from the wazir's department (diwan-i a’la-yi wizarat) to a completely new department, the diwdn-i

bandagan, with its own officials and headed by the ‘ arid-I bandagan-i khass. 114

Some of his slaves had been in Firuz Shah's service prior to his accession, like Malik Bashir, who

became carid-i mamalik with the style of cImad al-Mulik;115 or Malik *Dilan, who served the new sultan as

amir-i shikar, an office of increasing importance under a monarch who was so devoted to the chase;116 or

Malik Qabul, nicknamed Toraband, who became amir of Bada'un117 and is to be distinguished from a

namesake and fellow slave, Malik Qabul Qur'an-khwan, the amir-i majlis and muqtac of Samana.118

Subsequent purchases would have included Malik Ikhtiyar al-Din Mufarrij Sultani, the dawadar, who

became na'ib of the iqtac of Gujarat and later acquired the style of Farhat al-Mulk.119 By Firuz Shah's death, his slaves and their offspring constituted a major element in the aristocracy; we should be justified in

speaking of the creation of a new elite. The activities

112 Bihamadkhani, fols. 420a, 421a, 423, 424b, 425b, 432b (tr. Zaki, 27, 29, 32, 33, 34, 47).

TMS, 150. Cf. Digby, 'Iletmish or Iltutmish?', 57 n.l. For Juzjanis usage, see TN, I, 429 and n.3 (tr.

566, 567).

113 ‘Afif, 267-8, 269-70, 272: the figures are given at 270. 114 Ibid., 271.

115 TFS, 581. ‘Afif, 285, 436-7. TMS, 119.

116 TFS, 582. cAfif, 115, 318. Shokoohy, Rajasthan I, 62. On the importance of his office, see TFS,

600.

117 Ibid., 528. See also ‘Afif, 159-61. Bihamadkhani, fol. 417a: the text is slightly corrupt, reading

NWRABAD, and Zaki's tr. (23 and n.5) confuses him with Qabul Qur'an-khwan, as does TMS, 135, when

mentioning his appointment to Bada'un.

118 cAfif, 454-5. Bihamadkhanl, fol. 417a (tr. Zaki, 23). TMS, 134. He was sent against the

invading Mongols in 759/1358: ibid., 127.

119 Ibid., 133. He appears in epigraphs from 762 onwards: Desai, 'Khalji and Tughluq inscriptions

from Gujarat', 9-13, 19-21, 26-7, etc. (for a reconstruction of his career, see ibid., 13-14); idem, 'A fourteenth-century epitaph from Konkan', EIAPS (1965), 9-10.

~of these slaves under his successors would gravely undermine the stability of the empire.

Even discounting the amirs whom Firuz Shah had inherited from his cousin, however, there were

still several nobles of free stock. Zafar Khan (II), the muqta’ of Gujarat, was the son and successor of Zafar

Khan (I), whose full name, Taj al-Din Muhammad Lur Farsi, indicates that his family probably came from

south-west Persia.120 Malikzada Firuz (ancestor of the dynasty that ruled at Kalpi in the fifteenth century),

who held the extensive new shiqq of Firuzpur, was the son of Taj al-Din Turk, who had served Ghiyath al-

Din Tughluq Shah.121 The free maliks also comprised a group of Afghan amirs: Malik Ba'yyu, muqta’ of Bihar; Malik Khattab, appointed to the shiqq of Sambhal in 782/1380; and Malik Muhammad Shah, muqtac

of Tughluqpur in Etawa.122 Indian converts related to the sultan by marriage, too, found a place in the ranks

of the aristocracy. If we can believe a seventeenth-century historian of Gujarat, Sadharan, entitled Wajlh al-

Mulk, the ancestor of the independent sultans, was the brother of one of Firuz Shah's wives; he had

accompanied the sultan to Delhi and adopted Islam.123 By this time, lastly, leading figures among the local

princes enjoyed a place at court. After his campaign against Damrila, the sultan took its princes, the Jam

and his brother Banbhina, back to Delhi.124 By his death Uddharan, brother of the Tomara rai of Gwaliyor,

and *Sumer, the Chawhan rai of Etawa, were also both in attendance.125

Lineage and continuity

We would wish' to know more about the ancestry of most of the great nobles of the Khalji and Tughluqid periods mentioned in our sources. Isami tells us, for example, that Alp Khan, one of ‘Ala' al-

Din's early associates and subsequently governor of Gujarat, was of royal descent, and later alludes to the

illustrious ancestry of Baha' al-Din Garshasp, whose

120 TMS; 126; cf. also SFS, 91. Desai, 'Khalji and Tughluq inscriptions from Gujarat', 15-17.

121 Bihamadkhani, fol. 412b (tr. Zaki, 13-14); for the date, see TMS, 134, and on his background,

‘Afif, 480. For Taj al-DIn Turk, see TFS, 424.

122 Bayyu: Bihamadkhani, fol. 417a (tr. Zaki, 22, reading 'Babbu [?]'); TMS, 133 (BYR in error for

BYW); cf. also Z. A. Desai, 'Arabic and Persian inscriptions from the Indian Museum, Calcutta', ElAPS (1955-6), 6-8, for his epitaph of 753/1353 (here called Ibrahim Bayyu); Q. Ahmad, Corpus, 34-7 (nos. 11-

13). Khattab: TMS, 135. Muhammad Shah: Bihamadkhani, fols. 412a, 417a (tr. Zaki, 13, 23).

123 Sikandar b. Muhammad, alias (‘ urf) Manjhu, Mir'at-i Sikandarl, ed. S. C. Misra and

Muhammad Lutf al-Rahman (Baroda, 1961), 4-10; tr. in Sir E. C. Bayley, The local Muhammadan

dynasties. Gujarat (London, 1886), 67-70.

124 SFS, 93-4. cAfif, 247-8, 252-4. Riazul Islam, 'The rise of the Sammas in Sind', IC 22 (1948),

377-9.

125 ‘Afif, 281. TMS, 134. Bihamadkham, fol. 414b (tr. Zaki, 17). For the identification of these

princes, see Hodivala, Studies, I, 394-5; K. S. Lai, Twilight of the Sultanate, revised edn. (New Delhi, 1980), 6-7, n.35.

~mother was Tughluq's sister.126 But in neither case does he inform us who were the forebears of

the amir in question. On balance, our ignorance of lineage probably means that the aristocracy contained

fewer parvenus than might seem to have been the case, and we should not be unduly influenced by Barani's

evident obsession with birth as a qualification for office. Barani is in any case glaringly inconsistent, in that

he ignores the fact that the great noble families of Balaban's reign were descended from that sultan's fellow-

slaves, who could hardly be described as of good birth; and when impugning the birth of those who rose to

high office in 'Ala' al-Din's last years, he neglects to level the same charge at the upstart nobles of the early

part of the reign.127

Given the sudden and arbitrary manner in which amirs could be deprived of life and property and

their families disinherited, it is easy to ignore continuity. There were always grandees whose period of

service spanned different dynasties. Khwaja Jahan Ahmad b. Ayaz, wazir to Muhammad b. Tughluq, was

the son of cAla' al-Din Ayaz, kotwal of the Hisar-i Naw at Delhi (i.e. the new fortress of Siri) under

cAla'

al-DIn Khalji and kotwal of the capital in 720/1320, when he sent his son out with the keys to welcome

Ghiyath al-Din Tughluq; Ibn Battuta was told that the family was of Rumi origin.128 Malik Bashir Mu'izzi,

na'ib-i khass-hajib to Qutb al-Din Mubarak Shah, may well be identical with the Malik Bashir Sultani who

appears among the nobles of Mu'izz al-Din Kayqubad thirty years before.129 One of Qutb al-Din Mubarak

Shah's amirs was Aram Shah, the son of Malik Khurram Kuhijudi, who had served the Ghiyathids.130

The

father of Ahmad-i *ChhItam, cAla' al-Din's qirabeg, who with his sons held office under Qutb al-Din and Ghiyath al-Din Tughluq, was Balaban's slave, Malik *Buqubuq (above, p. 101).131 Malik Husam al-Din

Pindar Khalji, who served Qutb al-Din, received from Ghiyath al-Din the style of Qadr Khan, was sent at

the accession of Muhammad b. Tughluq to govern Lakhnawti, and was assassinated there on the outbreak

of insurrection in

126 Alp Khan: FS, 250, ki aslash bud az nutfa-yi shahryar (tr. 420). Baha' al-Din: ibid., 384, an

sipahdar-i wala-nasab (tr. 588).

127 I. Habib, 'Barani's theory', 107. Siddiqi, 'Nobility under the Khalji Sultans', 64-5.

128 IB, III, 144 (tr. Gibb, 617-18). For Ayaz, see TFS, 278; FS, 386-7 (tr. 590). The source cited by

A. M. Husain, The Rehla of Ibn Battuta (Baroda, 1953), 54 n.3 (and in turn in Gibb's tr., 655 n.131), which makes Ahmad b. Ayaz out to be of Indian origin, must be regarded as less trustworthy. HN, 447, 457,

mistakenly calls the father Ahmad and the son Muhammad.

129 TFS, 126 (reading BSYR, with BL ms., fol. 67b, for the YSR of the text). TMS, 83.

130 Ibid., 84.

131 Buqubuq: TFS, 40. Qirabeg: ibid., 331-2, 337, 379, 396, 424. FS, 287 (tr. 461), gives his name.

His sons: TFS, 409, 410. For the youngest, Badr al-Din Abu Bakr b. Ahmad, also named in an inscription

of 723/1323, see TFS, 379; G. H. Yazdani, 'Inscriptions in the tomb of Baba Arjun Shah, Petlad (Baroda

State)', E/M (1915-16), 16-18; ARIE (1975-6), 145 (no. D114).

~c. 1336.132 His father, Jamal al-Din Khalji, na'ib-i amir-i dad under both Balaban and Jalal al-

Din, had supported Jalal al-Din's sons in 695/1296 but was one of the very few maliks spared by ‘Ala' al-

Din in his purge of the Jalali nobles shortly afterwards.133 As we saw (p. 80), he may possibly be the Khalaj

chamberlain Jamal al-Din 'Ali who had acted as Balaban's agent on a mission to the Mongols.

The service of some noble families straddled three generations or more, though not always in a

strictly office-holding capacity. Barani, whose father, uncle and maternal grandfather all held office in the

latter half of the thirteenth century, seems never to have been more than.a boon-companion (nadim) to

Muhammad b. Tughluq. From what he says of the descendants of the thirteenth-century wazirs Junaydi and

Muhadhdhab al-Din, it appears that they were living as private citizens until the reign of Ghiyath al-Din

Tughluq, who restored them to favour, so that we find Husam al-Din Junaydi master-minding the assessment of the total revenue demand of the empire in FIruz Shah's reign, and his son Rukn al-DIn

Junaydi ('Junda') briefly serving as wazir to Abu Bakr Shah in 791/1389.134 The sayyid al-hujjdb Ma'ruf,

boon-companion to FIruz Shah, had been merely a military officer (pishwa) under Muhammad's amir

Sartiz; but his father was no less a figure than Wahid al-Din Qurayshi, na'ib-wazir and governor of Gujarat

for Qutb al-DIn Mubarak Shah.135 Malik Mahmud Beg, who held Sunnam and Samana under Muhammad

b. Tughluq and FIruz Shah successively, and on whom the latter conferred the title of Shir Khan, belonged

to a family from Bilahur that produced a series of office-holders proper. In view of Shir Khan's advanced

age (somewhat implausibly set at ninety by Barani), his father Rustam-i Yahya, muqtac of Bidar, must have

been promoted at some time during the Khaljl era.136 Shir Khan's own sons, Malik Abu Muslim and Malik

Sha,hin Beg, are later mentioned as officers of Firuz Shah.137 Perhaps the most striking instance of

continuity is provided by the genealogy of Dawar Malik, son of a sister, and also son-in-law, of Muhammad b. Tughluq: through his father Sadr al-DIn cArif, deputy chief

132 TFS, 379, 424 (with BYDAR in error; cf. BL ms., fols. 188b, 211a), 450, 454, 480. FS, 396 (tr.

601). TMS, 98.

133 TFS1 Bodleian ms., fol. 132a, calling him 'Jamal al-Din Khaljl, na'ib-i amir-i dad'; the

corresponding passage in TFS, 251, specifies that he was Qadr Khan's father, but calls him simply 'Amir

Jamal Khalji' (the printed text reads JMALY in error).

134 Ghiyath al-DIn Tughluq: ibid., All. Husam al-DIn: cAfif, 94, 460, 469-70, 481. Rukn al-Din:

ibid., 482; TMS, 143-4; and see also Hodivala, Studies, I, 391-2.

135 Sayyid al-Hujjab: ‘Afif, 445-6. Qurayshi: TFS, 397-8.

136 Ibid., 545, 583. TMS, 119, 120-1. Malik Rustam-i Yahya is found among Muhammad b.

Tughluq's maliks only in the BN ms. of TFS, Suppl. persan 251, fol. 282b. For the nisba Bilahuri, see IM,

106 (BLAHWDY in error); for Bilahur (modern Phillaur), on the right bank of the Sutlej, at 31° 1' N., 75° 48'

E., Punjab district gazetteers, XIVA. Jullundur (Lahore, 1904), 301. Shir Khan was dead by 765/1364,

when 'Malik Fakhr al-Dawla wa'l-DIn, son of Shir Khan Mahmud Beg', built a mosque at Patan: Desai,

'Khaljl and Tughluq inscriptions from Gujarat', 14-15.

137 TMS, 122. One of these may have been the builder of the above-mentioned mosque.

~qadi to ‘Ala' al-Din Khalji, he was a great-grandson of the chronicler Juzjani, chief qadi to two of

Iltutmish's sons and under Ghiyath al-Din Balaban.138

The survival of aristocratic families from one reign or dynasty to another was simply not

newsworthy in the way that the downfall of established amirs or the promotion of 'new men' always was; it

demanded less attention on the part of the chroniclers. Barani says in a few words that Ghiyath al-Din

Tughluq maintained in position the nobles of ‘Ala' al-Din KhaljI's reign; he devotes rather more space to

the predilection of Muhammad b. Tughluq for lowborn servitors. Yet Ghiyath al-Din, as we have seen, did

not simply nurture the aristocrats who represented the Khalji era: he also brought with him from Deopalpur,

and installed in office, members of his own retinue, men whose ability, courage and loyalty had been proven in his service over the past twenty years or so on the Mongol frontier. And in Muhammad's reign,

conversely, a good deal of evidence is to be found that old established families were still represented

among the office-holders.

We have little information about local aristocracies. The sources afford the occasional glance at a

local power-base - 'Ala' al-Din Khalji's at Kara, for instance, where his capacity for intrigue was

exacerbated by the influence of former associates of the Ghiyathid Malik Chhajju;139 or Malik Kafur's at

Rapri on the middle Yamuna and later at Deoglr.140 But it is not until the last decades of the fourteenth

century that Bihamadkhani's Ta'rikh-i Muhammadi enables us to trace what may truly be called 'local

history', in this case of the region around Kalpi and even then the ruling dynasty was an importation from

Delhi - the progeny of Malik Taj al-DIn Turk.

Otherwise the material that could have told us about Muslim notables in the provinces is meagre

indeed. Barani devotes some space to the families of sayyids resident in various towns of the Sultanate in cAla' al-Din's time, giving especial prominence to those of Bada'un, who served as qadis there. Two

members of the family, Sayyid Taj al-DIn and his nephew Sayyid Rukn al-DIn, attained the dignity of qadi

at Awadh and Kara respectively, and Barani says that he was privileged to meet them. He himself was

descended through his paternal grandmother from another distinguished clan, the sayyids of Kaithal.

Sayyids from Gardiz (presumably now dispersed around the empire) and those of Jajner and of Bhayana

are also mentioned.141 Members of these prestigious families emerge occasionally in the higher ranks of the

aristocracy. Malik Taj al-Din Ja’far, of the line of sayyids of Jajner, is listed among the maliks of ‘Ala' al-

Din and Qutb al-Din and subsequently became nd'ib-i 'ard and governor of Gujarat under

138 TFS, 351. TMS, 98. See also Hodivala, Studies, I, 309-10. 139 TFS, 224.

140 Ibid., 328, 333. Yazdani, 'Inscriptions of the Khalji Sultans of Delhi', 30.

141 Bada'un: TFS, 348-9. Kaithal: ibid., 349-50. Gardlz and Jajner: ibid., 350. Bhayana: ibid., 351.

~Ghiyath al-Din Tughluq.142 And on occasions these local lineages might find their fortunes

disrupted by royal violence. The sayyids of Kaithal paid dearly when one of their number, Sayyid Hasan,

kotwal of Madura, successfully revolted against Muhammad b. Tughluq in 734/1334 and (as Sultan Jalal

al-Din Ahsan Shah) founded the independent sultanate of Ma’bar: they were all massacred on the sultan's return from an abortive attempt to recover the province.143 Sayyid Hasan's son Ibrahim, keeper of the purse

(kharitadar) to Muhammad and governor of Hansi and Sarsati, was executed later on a charge of

conspiracy to revolt.144

In the provinces a rare shaft of light illuminates the resilience of distinguished local families even

over a stretch of a few centuries. We might think it highly improbable that any Indian locality in the reign

of Firuz Shah could preserve some connection with the Ghaznawid era. Yet an inscription from the Nagawr

region enshrines precisely that when it commemorates five brothers who fell in battle with the Hindus near

Bari Khatu in 761/1360.145 They bore the surname 'Bahalim', and hence must have belonged to the clan of

the powerful amir and founder of Nagawr, Muhammad Bahalim, who had rebelled against the Ghaznawid

Sultan Bahram Shah in 513/1119.146 It is a tantalizing thought that many more such venerable Muslim

aristocracies in the regions, beyond the horizons of our literary sources, may have survived every upheaval at Delhi.

142 Ibid., 240, 379, 424, 428; see 350 for his ancestry.

143 TMS, 106 (with incorrect date 742); and see 107 for the massacre of the sayyids. FS, 469 (tr.

705), for Sayyid Hasan's rank: N. Venkataramanyya, The early Muslim expansion in South India (Madras,

1942), 123-4 n.50 and 160, therefore argues that he did not govern the entire province, pace IB, III, 328-9

(tr. Gibb, 715). He is earlier found acting as na'ib to the governor of the Damoh region in 725/1325: B. D.

Verma, 'Inscriptions from the Central Museum, Nagpur', EIAPS (1955 6), 109-12; ARIE (1969-70), 84 (no.

D66).

144 IB, III, 337-9 (tr. Gibb, 718-19).

145 N. M. Ganam, 'An epitaph of six martyrs from Bari Khatu in Rajasthan', EIAPS (1973), 10-13.

146 TN, I, 242 (tr. 110); and see also AH, 378-81. Bosworth, Later Ghaznavids, 102-3.

~CHAPTER 10

An age of conquest

In chapter 7 an attempt was made to depict the constraints to which Muslim expansion was subject

in the thirteenth century. This is not to say that territories outside the control of the Delhi Sultan at that time were untouched by Islam. Muslim traders had been active in the maritime cities of the peninsula and in

Ceylon since the ninth century,1 and these regions retained close links with the Gulf: in the early thirteenth,

the khutba in different parts of 'Hind' was made in the name of the ruler of Fars.2 Kanbhaya (Cambay) in

Gujarat had its population of Muslim traders, scholars and lawyers for several decades before ‘Ala' al-Din

Khalji's forces first entered the country in 698-9/1299-1300. The presence of a flourishing Muslim

community in the twelfth and thirteenth centuries is demonstrated by the number • of epitaphs that have

come to light.3 These Muslims had their own prefect (hakim) on the eve of the Khalji invasion, and ‘Awfi

had been qadi there for a time in the early 1220s.4 It was the same in the far south. Ibn Battuta's description

of the Mulaybar (Malabar) coast demonstrates that communities of Muslims had settled in its- ports and

constructed mosques and hospices.5 At the time of his visit, there were 20,000 Muslims in the army of the

Hoysala king of Dvarasamudra; and a generation earlier ‘Ala' al-Din's invading army had encountered Muslims among the forces of

1 Genevieve Bouchon, 'Quelques aspects de l'islamisation des regions maritimes de l'lnde a

l'epoque medievale (XlIe-XVIe s.)', in Gaborieau, Islam et societe en Asie du sud, 29-36. Wink, Al-Hind, I,

67-83.

2 Ibn Zarkub, ed. KarimI, 56/ed. Jawadi, 80.

3 Z. A. Desai, 'Arabic inscriptions of the Rajput period from Gujarat', ElAPS (1961), 1-24; idem,

'Early Kufi epitaphs from Bhadreswar in Gujarat', EIAPS (1965), 1-8; ARIE (1961-2), 33, 179 (nos. D22-29).

4 Desai, 'Khalji and Tughluq inscriptions from Gujarat', 3-4. For ‘Awfi's position at Kanbhaya,

which was misunderstood by Nizamu'd-din, Introduction, 14, see Hodivala, Studies, I, 171, and II, 44.

5 IB, IV, 71-103 (tr. Gibb and Beckingham, 805-18). See generally G. F. Hourani, Arab seafaring

in the Indian Ocean (Princeton, 1951); Stephen F. Dale, Islamic society on the South Asian frontier: the

Mappilas of Malabar 1498-1922 (Oxford, 1980), chapter 1; Wink, Al-Hind, II, 276-80.

~the Pandya kingdom of Ma'bar.6 Muslim rule, then as now, has never kept pace with the Muslim

diaspora.

The decisive forward thrust is associated with the era of cAla' al-Din Khalji (695-715/1296-1316).

His seizure of the throne had been made possible by a raid of unprecedented audacity upon the distant

Yadava kingdom of Deogir; and his reign as sultan was characterized by equally ambitious campaigns

against independent Hindu powers in Rajasthan, in Malwa and south of the Narbada. Yet for all that we

may feel confronted at this juncture by some kind of quantum leap in the process of expansion, we should

beware of identifying every triumph as a landmark. Some of ‘Ala' al-Din's campaigns were simply an

extension of the activity of thirteenth-century sultans: his reduction of Ranthanbor, for instance,

representing the recovery of a stronghold that had twice previously been in Muslim hands.

Nor are we necessarily faced everywhere with a new stage of conquest and absorption. True, the

military reach of the Sultanate had been dramatically extended; Hindu fortresses which had hitherto yielded merely spasmodic tribute now became the seat of a Muslim governor or muqta’. But the push to the south

often did no more than replicate the pattern discernible in the north over the previous hundred years or so.

In Baranis account of the first successful invasion of Tilang, ‘Ala' al-Din instructs his general, Malik Kafur,

not to make any effort to take the fortress of Arangal or to overthrow its king: if treasure, jewels, elephants

and horses were offered, and tribute guaranteed for future years, he was to reach a rapid settlement.7

Frequently the submission of a prince was accepted by the sultan or his representatives with the

characteristic reassuring gesture of the hand on the back, extended to the envoys of Rudradeva of Tilang by

Kafur and to the rai of Nagarkot by Firuz Shah.8 There is the same pronounced gap between the ideal -

exercise of immediate administrative control over the entire territory of a defunct Hindu kingdom - and a

reality that resided in inevitable compromise with local powers or- confined direct Muslim rule to a handful

of major strongpoints. As we shall see, the last years of the Khalji era would witness the beginnings of a

more forward policy.

6 Hoysalas: IB, IV, 195-6 (tr. Gibb and Beckingham, 861). Pandyas: KF, 149; DR, 72.

7 TFS, 327: a line is omitted here which is found in BL ms., fol. 162a, ki dar giriftan-i hisar-iI

Arankal wa-bar andakhtan-i rai Laddar Diw mubalighi makuni wa-agar DIW RAI-YI

Arankal khizana ...

8KF, 104,’AFif, 189,244. For Nagarkot, see also L. S. Chandel, 'References to Kangra and Sirmur

in the early medieval Persian sources', in his Early medieval state (a study of Delhi Sultanate) (New Delhi,

1989), 104 and n.39; further references in Hodivala, Studies, I, 321.

~The campaigns in northern India

Gujarat

Mahmud of Ghazna had sacked the temple city of Somnath on the coast and the wealthy entrepot

of Nahrwala (Anhilwara; modern Patan); and in 593/1197 Nahrwala had again been looted by Qutb al-Din

Aybeg. But no Muslim attacks on Gujarat are recorded thereafter, either upon the Chaulukyas or upon their

Vaghela kinsmen who succeeded them around 1242, until almost the end of the century. Then in 698/1299

'Ala' al-Din sen thibrother Ulugh Khan and the wazir Nusrat Khan against the Vaghela kingdom. The

principal goal of the expedition may have been to sack Somnath, doubtless in conscious emulation of Mahmud of Ghazna. The Vaghela king Karnadeva seems to have disputed the passage of the Muslim army

as it neared Gujarat, and was defeated and fled south-east to Baglana (in the Nasik region). Somnath and

Nahrwala were plundered, and Nusrat Khan sacked Kanbhaya, probably in Dhu'l-Hijja/September 1299. A

Jain inscription tells us that although Satyapura (Sachor) was saved by a miracle the sultan's forces overran

the Kathiawad peninsula. After this, ‘Ala’al-Din's generals withdrew to Delhi with an enormous booty,

their progress interrupted only by an abortive mutiny on the part of some neo-Muslim Mongol

commanders.9

Contrary to the impression sometimes given, this campaign did not entail a Muslim conquest or

the definitive overthrow of the Vaghelas. Barani's account of the campaign includes the misleading

statement that Karnadeva sought asylum at Deogir with the Yadava king Ramadeva. In fact, this occurred a

few years later. A bilingual inscription of 704/1304 shows Karnadeva established at Vadodara (Baroda), on the eastern marches of his kingdom, and flanked, incidentally, by two of the neo-Muslim Mongol amirs

who had deserted the Delhi army.10 Moreover, 'IsamI describes a second invasion of Gujarat, effected by

the qirabeg Ahmad-i *Chhitam. It seems to

9 See generally S. C. Misra, The rise of Muslim power in Gujarat, 2nd edn (New Delhi, 1982), 61-

4; Lal, History of the Khaljis, 67-73; HN, 334-6. The date of the farman ordering the carid to prepare for the

expedition is given as Wed. 20 Jumada I 698 in a couplet in KF, 47 (Lal, History of the Khaljis, 68,

mistakes this for the actual date of departure): since the 20th was in reality a Monday (23 February 1299),

Hodivala (Studies, I, 248-9) assumed that the year intended was 697. But the correct date is in fact found in

a chronogram a few lines earlier: 22 Jumada I 698/25 Feb. 1299 (KF, 46-7). The month Dhu'l-Hijja given

by Wassaf, 447, probably refers to the attack on Kanbhaya. For the date of the sack of Somnath (June 1299), see D. B. Diskalkar, 'Inscriptions of Kathiawad', NIA 1 (1938-9), 695. The campaign must have

extended into the year V.s. 1356 (1299-1300) given in Jinaprabha's Tirthakalpa almost thirty years later: G.

Biihler, 'A Jaina account of the end of the Vaghelas of Gujarat', I4 26 ( 897), 194-5.

10 TFS 251. Z. A. Desai, 'A Persian-Sanskrit inscription of Kama Deva Vaghela of Gujarat', EIAPS

(1975), 13-20; ARIE (1980-1), 6-7, 123-4 (no. B98). See FS, 255 (tr. 425), for the flight of the two Mongol

amirs to Karnadeva.

~have been this expedition which sacked Sachor in 1310; and it was only now that Karnadeva fled

to the Deccan and hence to Tilang and that ‘Ala’al-Din appointed the first Muslim governor in the person

of Alp Khan.11

Even after this second assault, however, Muslim domination of Gujarat remained patchy: it may,

in fact, have been undermined by the revolt of the provincial garrison troops that followed Alp Khan's

execution (p. 177). In Gujarat, as elsewhere, Ibn Battuta notices 'rebels who inhabit inaccessible retreats in

the mountains'.12 An early seventeenth-century author asserts that ‘Ala' al-Din established the light of Islam

only in the territories lying east of a line drawn from Nahrwala to Bharuch, and that the eradication of

pagan practices in the outlying parts (atraf-u jawanib) dated from the period of the independent sultans

from the fifteenth century onwards.13 This may not do justice to the Tughluqid era, but it is true that local

political conditions varied widely. Epigraphical evidence of Muslim rule down to Muhammad b. Tughluq's

reign is concentrated in the east, in Petlad, Patan, Bharuch and Kanbhaya.14 Nawsari is found as an iqta' by

725/1325.15 But in 745/1344-5 Nanadeva, the chief (muqaddam) of Salher (Salir) and Mulher (Malir) in

Baglana, appears to have been virtually independent.16 And much of the Kathiawad peninsula lay outside the writ of the sultan's governors. Gandhar's ruler paid tribute to Delhi at the time of Ibn Battuta's visit in

the early 1340s; yet the Hindu ruler of Gogha (Quqa), he tells us, who had at one time professed allegiance

to Delhi, did so no longer.17 Members of the Vaghela dynasty continued to rule at Dandahidesa under the

overlordship of Delhi and, later still, in. subordination to the independent Gujarat sultans.18

At Vamanathali

(Vanthali), whose rana Mamdalikka had been chastised by Ulugh Khan in the 1299 campaign, his dynasty,

the Chudasamas, contrived to extend their power over much of the Girnar (Junagarh) region.19 When he

entered Gujarat in

11

Misra, Muslim power, 64-6, though dating this second invasion in 704/1304-5. FS, 286-8 (tr.

461-3). For the sack of Satyapura (Sachor) in V.s. 1366/1310, see Buhler, 'Jaina account', 195. That Karnadeva was not welcomed in Deogir suggests that his flight postdated the Yadava king's submission to

Delhi in 706/1307 (below).

12 IB, III, 245 (tr. Gibb, 672).

13 Sikandar 'Manjhu', Mir'at-i Sikandari, 42 (tr. Bayley, 97).

14 Petlad: Misra, Muslim power, 67; ARIE (1975-6), 145 (no. D114, 723/1323; no. D115, 713/

1313). Patan: Z. A. Desai, 'An early fourteenth century epigraph from Gujarat', ElAPS (1970), 13-15

(715/1315). Bharuch: M. Nazim, 'Inscriptions from the Bombay Presidency', EIM (1933-4), Supplement,

25-6 (721/1321-2), 27 (726/1326). Kanbhaya: A. M. Husain, 'Six inscriptions of Sultan Muhammad bin

Tughluq Shah', ElAPS (1957-8), 29-34 (725/ 1325); ARIE (1973-4), 143 (no. D80, 734/1334). 15 TMS, 98. Q. M. Moneer, 'Two unpublished inscriptions of the time of Sultan Muhammad bin

Tughluq', EIM (1939-40), 24-6.

16 TFS, 512, calling him 'Man Diw'; for the identification, see Hodivala, Studies, I, 299.

17 IB, IV, 58, 59, 61 (tr. Gibb and Beckingham, 799, 800, 801).

18 Ray, Dynastic history, 1046 n.l.

19 For Mamdalikka, see Buhler, 'Jaina account', 194; for the epigraphy of the dynasty, Diskalkar,

'Inscriptions of Kathiawad', 576-90, etc. (especially 578-9).

~pursuit of the rebel Taghai in 1349, Muhammad b. Tughluq arrested 'Kanhgar', rana of Ginar,

and imposed his own revenue-collectors on the region.20 It is indeed possible that his three-year stay in

Gujarat - the first visit by a reigning Delhi Sultan - brought about an intensification of control over the

province.

Rajasthan and Malwa

In the thirteenth century Ranthanbor had been the objective of several campaigns from Delhi. Its

raja, Hammlradeva, who is described by 'Isami as a friend of ‘Ala' al-Din, had nevertheless created a casus

belli by giving shelter to certain of the Mongol amirs who mutinied during the first Gujarat campaign;21 and following the great Mongol invasion by Qutlugh Qocha in 1299-1300, the sultan sent Ulugh Khan, then

muqta' of Bhayana, and Nusrat Khan, muqta’ of Kara, with the army of 'Hindustan' to attack the fortress.

The two generals took Jhayin, but during their investment of Ranthanbor Nusrat Khan was mortally

wounded. The sultan therefore set out in person to take charge of the siege operations. According to Amir

Khusraw, who dates the beginning of the siege in Rajab 700/March 1301, Ranthanbor was taken on 3

Dhu'l-Qacda/11 July.22 Hammlradeva and his neo-Muslim guests fell in the fighting. Ranthanbor and its

dependencies were conferred on Ulugh Khan, who died, however, within a few months.23 It is a measure of

the firmness of the Muslim hold here that under his successor, Malik ‘Izz al-Din *Bura Khan, Jhayin,

renamed Shahr-i Naw ('New City'), could be subjected to the same system for collection of the land-tax

(kharaj) that obtained in the heartlands of the Sultanate (see below, pp. 242-4).24

During the outward march of Ulugh Khan and Nusrat Khan towards Gujarat in 698/1299,

Samarasimha, raja of Chitor, had protected his kingdom by paying tribute.25 It seems that he subsequently

reneged on his submission, for in Jumada II 702/January 1303 ‘Ala' al-Din in person set out for Chitor.

Barani makes only fleeting mention of this campaign, describing the siege as brief. But Khusraw says that

the place capitulated on 11 Muharram 703/26 August, and that the raja surrendered to 'Ala' al-DIn.26

20 TFS, 521, 523; for the identification of Kanhgar', see Hodivala, Studies, I, 302-3.

21

TFS, 283. FS, 255, 271-3 (tr. 425, 446-7). TMS, 11.

22 KF, 51, 54, for these dates, on which see Hodivala, Studies, I, 249. On the Ranthanbor

campaign, see Lal, History of the Khaljis, 83-6, 89, 93-6.

23 DR, 66. TFS, 283; cf. also 299 for Ulugh Khan's death.

24 Ibid., 288, 299, 306. For the renaming of Jhayin, see KF, 54; more references in Gupta, 'Jhain of

the Delhi Sultanate'.

25Btihler, 'Jaina account', 194. The phrasing is ambiguous: see Lal, History of the Khaljis, 69.

26 KF, 60, 61-2, 63. TFS, 299. FS, 281 (tr. 456), calling the raja Samarasimha in error for the

latter's son and successor Ratan Singh. The date is discussed by Hodivala, Studies, I, 250; for the siege, see Lal, History of the Khaljis, 99-102.

~Chitor was renamed Khidrabad in honour of the sultan's son and heir-presumptive, Khidr Khan,

who became its nominal governor.27 'Isami makes it clear, however, that the administration was entrusted to

'Ala' al-Din's slave Malik Shahln.28 The story found in Sanskrit epic and purveyed also by Firishta,

according to which after 'Ala' al-Din's death the fortress was first occupied by a brother of the raja of Jalor

and then passed into the hands of the rajas of Sisodia for two centuries, is not borne out by epigraphical

evidence, which shows that Chitor was still ruled by governors sent from Delhi in the reigns of the first two

Tughluqid sultans.29

According to Amir Khusraw, the Delhi forces had been investing Siwana (Sevana) for five or six years before it fell.30 Be that as it may, after 'Ala' al-Din took charge of the investment the fortress was

taken in Rabi' I 708/ August-September 1308 and the raja 'Satal Deo' was killed. Siwana, renamed

Khayrabad, was conferred on Malik Kamal al-Din 'Gurg',31 who is also credited by Sirhindi with the

capture of Jalor and the overthrow of its raja, 'Kanhar Deo' (Kanhadadeva, son and successor of

Samantasimha) around the same time. Barani makes only a passing allusion to the incorporation of both

places within the sultan's dominions, mentioning neither campaign; but it seems that Jalor fell in 1311 to

the same army that had sacked Sachor in the previous year.32 It is clear from inscriptions of 1318 and 1323

that Jalor remained under Muslim rule into the Tughluqid era.33

In the course of his Ranthanbor campaign of 700/1301, the Delhi Sultan's forces had overrun 'the

territory (wilayat) of Jhayin as far as the frontier of Dhar'.34 But it was not until after the fall of Chitor that

'Ala' al-Din determined, in the words of Amir Khusraw, to 'seize the kingdoms of the southern rais'. In 705/1305 his army duly advanced into the Paramara kingdom of Malwa. The Delhi forces first defeated a

potentate named 'Koka Pradhan', whom Khusraw calls a 'wazir' more powerful even than the rai himself

and who was at loggerheads with the king, 'Mahlak Deo'.

27 KF, 63-4. DR, 67. TMS, 11.

28 FS, 281, 282 (tr. 456, 457). According to TFS, 323, the governor was Malik Abu Muhammad,

who is otherwise unknown.

29 Z. A. Desai, 'Inscriptions from the Victoria Hall Museum, Udaipur', EIAPS (1955-6), 67-70. Cf.

Lal, History of the Khaljis, 110-12; HN, 371. 30 DR, 69.

31

KF, 68-72. For another account of the campaign, where the raja is called 'Sital', see FS, 315-17

(tr. 492-4). For Kamal al-Din's nickname, see Hodivala, Studies, I, 251.

32 TMS, 78, but dating the fall of both Siwana and Jalor in 700, which must be too early and

clashes with the testimony of Amir Khusraw. TFS, 323. Bhandarkar, 'Chahamanas of Marwar', 77-8, for the

date and the identification of 'Kanhar' (KSTMR in the printed text of TMS, and KTHR in one ms.); also Lal,

History of the Khaljis, 118-19. 33 1318: Desai, 'Jalor ‘Idgah inscription', correcting the earlier reading of G. H. Yazdani,

'Inscription of Mubarak Shah Khalji from Jalor, Jodhpur State', ElM (1935-6), 49-50. 1323: M. A.

Chaghtai, 'Some inscriptions from Jodhpur State, Rajputana', ElM (1949-50), 32. Shokoohy, Rajasthan I,

45-7.

34 TFS, 277.

~Then ‘Ayn al-Mulk Multani was sent against Mandu, where he besieged the raja in person. The

place was taken on 5 Jumada 1/23 November 1305, and ‘Ayn al-Mulk, on whom the sultan had already

conferred Malwa, was rewarded with the further grant of Mandu.35 Barani says nothing about the conquest

of Malwa, but confirms that during ‘Ala' al-Din's reign Mandal-khur, Dhar, Ujjain, Mandugarh (Mandu), 'Ala'Ipur, Chanderi and Erach were all allotted to governors (walls) and muqta's.36 Precisely when most of

these places were taken is unknown. Erach, renamed Sultanpur, was in Muslim hands by 709/1309, when

Malik Kafur halted there for five days en route for Arangal.37 Chanderi first appears as an iqta' in 711/1312

(p. 174 above). An inscription of 1310 at Udayapura (in the present-day Vidisha district) reveals that the

Paramara dynasty survived here in the northeastern part of the country; but in 739/1338 the inscription on a

new mosque testified to the sovereignty of Muhammad b. Tughluq.38

The Chandellas of Jejakabhukti (Bundelkhand) were in all likelihood subdued at some point

during the campaigns in Malwa, since an inscription of 1309 in a village near Bamhni acknowledges ‘Ala'

al-Din's sovereignty, where only five years before a feudatory of Hammiravarman had been named. At any

rate, an epigraph of 1315 accords the obscure king Viravarman II only a shadow of the titles borne by his predecessors.39 A consequence of this, presumably, was the capture of Mahoba, although we have no

evidence of Muslim occupation prior to the construction of a mosque there in 722/1322, during the reign of

Ghiyath al-Din Tughluq.40 Further south, the Pratiharas appear finally to have been subjugated either by

Tughluq or by his successor, to judge from inscriptions of 1325-42 found in the Damoh and Jabalpur

districts.41 Beyond this region lay Gondhiyana (Gondwana), which Muhammad b. Tughluq penetrated in c.

1326, on his way back from the Deccan: Nag Nayak, 'chief of the Kolis', yielded after a lengthy siege of his

stronghold, but we do not know for how long he remained submissive and the history of this immense tract

is obscure.42

35 DR, 67-8. Hodivala, Studies, I, 249-50. For the grants to ‘Ayn al-Mulk, see KF, 56, 59, with the

date of the fall of Mandu (wrongly given in TMS, 77, as 700); DR, 69.

36 TFS 323; for Chanderi, see also ibid., 328. ‘Ala'ipur must be ‘Alapur, near Gwaliyor: IB, IV,

31 (tr. Gibb and Beckingham, 786).

37 KF, 75. On Erach/Sultanpur, on the S. bank of the Betwa, at 25° 47' N., 79° 9' E., see Hodivala,

Studies, I, 252-3, and DGUP, XXIV. Jhansi, 254-6.

38 ARIE (1961-2), 169 (no. C1637); also Ray, Dynastic history, 905-6, 908. For the mosque, see

ARIE (1964-5), 23, 145 (nos. D77-78).

39 P. Prasad (ed.), Sanskrit inscriptions, xviii-xix and 156-8. R. K. Dikshit, Candellas, 177-8,

citing Rai Bahadur Hiralal, 'Mahoba plates of Paramardi-deva: (Vikrama-) samvat 1230', EI16(1921-2), 9-15.

40 Z. A. Desai, 'Two inscriptions of Ghiyathu'd-Din Tughluq from Uttar Pradesh', EIAPS (1966),

23-6.

41 Verma, 'Inscriptions from the Central Museum, Nagpur'. ARIE (1962-3), 96 (no. B430). ARIE

(1967-8), 6, 27 (no. B108). ARIE (1969-70), 84 (no. D66).

2 FS, 432-3 (tr. 659-61): following the suppression of Garshasp's revolt.

~The Doab, Awadh and beyond

‘Ala' al-Din appears also to have tightened his grip on the regions east of Delhi, though we are ill

informed about both the details and the chronology. In the southern Doab, Jajmaw was under Muslim

occupation by 706/ 1307.43 The appearance of Rapri, on the Yamuna, as an iqta’ by 709/1309; an

inscription of ‘Ala" al-Din's time at Mathura, dating from soon after the first invasion of Gujarat; and the

emergence of Gwaliyor by the end of the reign as a place of confinement for important prisoners of state -

all this throws a faint light on the steady growth of the sultan's authority in regions where the government's

hold in the thirteenth century had been tenuous.44 More arresting are Barani's claims that under ‘Ala' al-Din

Katehr was subjected, like traditionally less recalcitrant districts in the heartlands of the Sultanate, to the

land-tax on the basis of measurement (see below, p. 243), and that an advance base like Kabar could be

incorporated in the crown lands (khalisa).45

The early Tughluqid period is notable in particular for an intensification of Muslim settlement in

the fertile region of Awadh. Iqta 's appeared here -Dalmaw, Bangarmaw, Lakhnaw (Lucknow) and Sandila

- which as far as we know had not existed in the Khalji era.46 From the beginning of Ghiyath al-Din

Tughluq's reign new strongpoints too were being constructed. The fortress at Zafarabad (later renamed

Jawnpur), for instance, which was conferred as iqta’ on the sultan's adopted son Tatar Khan, had been

completed in Rabi' I 721/April 1321 by Malik Mall, who also left an inscription dated Muharram of that

year/January 1321 in the Allahabad district.47 Possibly this burst of activity was designed as a prelude to the

sultan's intervention in the independent sultanate of Bengal in 724/1324, when he reinstated one of the two

sons of Shams al-Din Firuz (above, p. 95), Nasir al-Din, at Lakhnawti as his subordinate and replaced the

other, Ghiyath al-Din Bahadur 'Bura', at Sunarga'un, with his own 43 W. H. Siddiqi and Z. A. Desai, 'Khaljl and Tughluq inscriptions from Uttar Pradesh', EIAPS

(1964), 3-4.

44 Rapri: TFS, 328; and see Yazdani, 'Inscriptions of the Khaljl Sultans of Delhi', 30, and

Hodivala, Studies, I, 281. Mathura: Khan Bahadur Zafar Hasan, 'An inscription of ‘Alau'd-Din Khaljl

recently discovered at Muttra', EIM (1937-8), 59-61. Gwaliyor: TFS, 368; TMS, 72 (claiming that Ahmad-i

Chap and Alughu were incarcerated there at the very beginning of ‘Ala' al-Din's reign, although this is at

variance with TFS and could well be an error); IB, III, 188, 333 (tr. Gibb, 642, 717).

45 TFS, 288; for Kabar, see ibid., 323-4.

46 IB, III, 342, 349 (tr. Gibb, 721, 724). TMS, 93. Troops from the iqta’ of Bangarmaw participated

in the Tilang campaign of 721/1321: TFS1, Bodleian ms., fol. 183a/Digby Coll. ms., fol. 154b.

47 Desai, 'Two inscriptions of Ghiyathu'd-Din Tughluq from Uttar Pradesh', 19-23. G. H. Yazdani,

'Inscription of Ghiyathu'd-DIn Tughluq from Asrawa Khurd near Allahabad', EIM (1937-8), 6-7. For the

grant of Zafarabad, see TFS, 428, 451.

~officers.48 This was the first time the authority of the Delhi monarch had been

recognized in Bengal since the death of Balaban.

Nepal is said to have acknowledged ‘Ala' al-Din Khalji's overlord-ship,49 and the sixteenth-century writer Mulla Taqiya alleges that he imposed tribute on the raja of Tirhut. This seems to be corroborated by

the earlier recension of Barani' s history, which suggests that Tirhut was already supplying troops in

702/1302-3 for that sultan's ill-starred expedition to Tilang.50 But the raja must have asserted his autonomy

following cAla' al-Din's death, for Tirhut was raided during Qutb al-Din's reign by Malik Kafur the

muhrdar (not to be confused with the late Malik Na'ib), who extorted tribute from him. Only a few years

later, while returning from his Bengal campaign in 724/1324, Ghiyath al-Din Tughluq headed an attack on

Tirhut of which the fullest account is given by Ikhtisan-i Dabir, an eye-witness. The raja, Harisimhadeva,

fled to Nepal, and his capital fell to the Delhi forces. Isami tells us that Ahmad b. *Tulabugha was left there

when the sultan returned to Delhi.51

Barani counted Tirhut as a province subject to Muhammad b. Tughluq

a few years later, and coins were struck in his name at 'Tughluqpur, alias (curf) Tirhut' from at least 731/1330-1.52

Beyond the Narbada

The Deccan, Tilang (Telingana) and Kampila

At the time of ‘Ala' al-Din's raid on Deoglr in 695/1296, the Yadava king Ramadeva had

undertaken to pay regular tribute. But at some point -perhaps in reaction to the Delhi forces' unsuccessful

campaign against Tilang in 702/1302-3 - he neglected to do so, and in 706/1306-7 ‘Ala' al-Din sent his

favourite, the Malik Na'ib Kafur 'Hazardinari', against Deoglr. On 19 Ramadan/25 March 1307 Ramadeva's

army was defeated and he himself captured. ‘Ala' al-Din detained him in Delhi for about six months,

treating him kindly before sending him back to his capital as a subordinate ruler, with the title rai-yi rayan ('rai of rais') and a chatr. BaranI observes that Ramadeva remained submissive for the rest of his life;53 and

when in

48 For this campaign, see Husain, Tughluq dynasty, 74-6. TFS1, Bodleian ms., fols. 184b-185a/

Digby Coll. ms., fols. 155b-156a, furnishes a slightly fuller account than other sources.

49 Luciano Petech, Mediaeval Nepal (c. 750-1480), Serie Orientale Roma, X (Rome, 1958), 103-4.

50 Hasan Nishat Ansari, 'Political history of Bihar under the Khaljis (A.D. 1290-1320/A.H. 690-

720)', JBRS 54 (1968), 260-3. TFS1, Digby Coll. ms., fol. 113a.

51 Ikhtisan, Basatin, fols. lOa-llb (tr. Askari, 'Historical value', 11-12). FS, 365, 416-18 (tr. 564,

628-30): the editor, Usha, points out that some lines are omitted from the account of Tughluq's campaign

here. For Harisimha's flight, see Petech, Mediaeval Nepal, 111-13.

52 TFS, 467. CMSD, 117 (no. 478), 140 (nos. 579-81).

53 TFS, 326. A more detailed account in KF, 64-8, with the date, on which see Hodivala, Studies, I,

250.

~710/1310-11 Kafur arrived at Deogir en route to attack Ma'bar, Ramadeva was not only

assiduous in furnishing provisions and reinforcements but ordered a subordinate rai to guide the Delhi army

on to Dvarasa-mudra.54 The route to Tilang through Deogir was safer than that by way of Sirpur as taken by Kafur in 709/1309, so that as ‘Isami - himself an inhabitant of the Deccan - recognized, the possession of

an advance base here was essential to campaigns elsewhere in the south.55 For a time Ramadeva's

compliance furnished the Delhi armies with just such a base; only after his death, when hostile elements

took control of the Yadava kingdom, was it necessary for ‘Ala' al-Din and, later, Qutb al-Din Mubarak

Shah to annex Deogir.

After Ramadeva's death, towards the end of ‘Ala' al-Din's reign, his son Singhanadeva headed a

reaction and had to be quelled by yet another expedition under Kafur, at whose approach he fled into the

hills. Kafur, who was appointed as governor and who is duly said to have demanded the account-books

(jara 'id) from the clerks (ahl-i qalam), was under instructions to levy taxes (mat) on the cultivators and to

build mosques.56 From 714/ 1314-15 coins were being struck in ‘Ala' al-Din's name at the Deogir mint.57 It is therefore clear that these operations by Kafur represent the first attempt at annexation of the Yadava

kingdom, a development which has sometimes been placed in the reign of Qutb al-Din Mubarak Shah. But

with ‘Ala' al-Din's final illness and death, the bonds between Deogir and the capital slackened. The sultan

recalled Kafur, who is subsequently said to have ordered his deputy ‘Ayn al-Mulk to bring the Muslim

inhabitants of Deogir to Delhi.58

Qutb al-Din's march south in 717/1317, according to Amir Khusraw, brought him the submission

of 'all the rais' except 'Raghu', deputy and minister to the late Ramadeva, who raised an army but was

defeated by the sultan's favourite Khusraw Khan and fled. On his way to rejoin the sultan, Khusraw Khan

also defeated and executed 'Harpal Deo', Ramadeva's son-in-law and a member of the defunct Chalukya dynasty formerly ruling in

54 KF, 122-4, 126. Ramadeva was not dead by this time, as alleged in TFS, 333: Venkatar-

amanyya, Early Muslim expansion, 50-1 n.88; Lal, History of the Khaljis, 245-6; later (ibid., 255) Lal dates

his death in 1312-13. Despite TFS, 328-9, Kafur advanced on Tilang in 1309, not by way of Deogir, but via

Basiragarh (variant reading for the BYJAGRHH of the text: this is Wairagarh in the Chandrapur district) and

Sarbar (i.e. Sirpur): KF, 80; Hodivala, Studies, I, 254-5; also Joshi and Husain, 'Khaljis and Tughluqs in the

Deccan', 45 (though stating that Ramadeva placed troops at his disposal).

55 FS, 360 (tr. 558). P. M. Joshi, 'Historical geography of medieval Deccan', in Sherwani and Joshi

(eds.), History of medieval Deccan, I, 12.

56 FS, 333-6 (tr. 513-16), is the sole source for this episode; see Lal, History of the Khaljis, 255-7.

Work on one mosque, at Naltawar in the Bijapur district, was completed in 715/ 1316: G. H. Yazdani, 'An

inscription of cAla'-u-din Khalji from Rakkasgi in the Bijapur district', EIM (1927-8), 16-17.

57 CMSD, 89 (no. 305C), 91 (nos. 321-2).

58 TFS, 368. FS, 336 (tr. 516), links Kafur's recall with the festivities for the marriage of the

sultan's son Shadi Khan; see ibid., 347-8 (tr. 528-9) for ‘Ayn al-Mulk.

~Kalyani, who had been entrusted with authority in the region but had risen in revolt.59 Other

sources make no mention of 'Raghu' and speak as if 'Harpal Deo' was the principal antagonist. cIsami's account suggests additional motives for the campaign, saying that Qutb al-Din was able to lay hands on the

wealth amassed in the region by Malik Kafur and making Harpal out to be a former confederate of the late

na'ib.60 Deogir, temporarily renamed Qutbabad in the sultan's honour, again became a mint and was

provided with an administration in the form of a wazir and revenue officials; the territory was apportioned

among muqta's.61 Several years later, in 1333-4, a certain Melugideva, son of Singhanadeva, built a temple

in the Dhule region and named Muhammad b. Tughluq as his sovereign: if this figure is indeed Ramadeva's

grandson, the Yadavas had lingered on as the sultan's subordinates.62

In the course of his last Deccan expedition, Malik Na'ib Kafur had briefly raided the kingdom of

Kampila, which had been founded in the latter part of the thirteenth century. Profiting from the collapse of

the Yadavas in the second decade of the fourteenth, its rulers had extended their authority over the modern

districts of Bellary, Chitaldrug, Raichur and Dharwad and established Kumta and Husdurg (Anegondi) as their two principal centres.63 Kafur ravaged the furthest parts of the territory and advanced as far as

Kumta.64 This same region may have been attacked by the future sultan Muhammad b. Tughluq following

his second Tilang expedition, since 'Isami refers obliquely to the reduction of GuttI (embracing parts of the

Anantapur and Bellary districts) and Kunti (Kumta?).65 But the conquest of the kingdom was deferred until

c. 1327, when the raja refused to surrender the rebel Baha' al-Din Garshasp to Muhammad b. Tughluq's

forces, which took Kumta by storm. Husdurg, whither Garshasp and his host fled, was taken in turn;

Garshasp escaped to Dvarasamudra, but the raja of Kampila was killed in the fighting.66 Kampila was now

subjected to the sultan's overlordship and Barani includes it in his list of

59 NS, 62-73, 195-202; see ibid., 196-7, for the earlier commission to Harpal. Joshi and Husain,

'Khaljis and Tughluqs', 50. TFS1 Bodleian ms., fol.l67a/Digby Coll. ms, fol. 143a, dates this expedition in 718/1318-19.

60 FS, 360-1 (tr. 558-9). TFS, 389-90, likewise mentions only 'Harpal'.

61 Ibid., 390. For coins of Qutb al-DIn from 'Qutbabad', see CMSD, 98 (no. 374A). See further

HN, 434-5; Joshi and Husain, 'Khaljis and Tughluqs', 51.

62 ARIE (1962-3), 24-5, 132 (no. B744).

63 See M. H. Rama Sharma, 'The kingdom of Kampila', Journal of the Bombay Historical Society

2 (1929), 201-8; K. A. Nilakanta Sastri and N. Venkataramanyya (eds.), Further sources of Vijayanagara

history (Madras, 1946, 3 vols.), I, 9-21; Venkataramanyya, Early Muslim expansion, 74-5.

64 FS, 335-6 (tr. 515-16).

65bid., 31 (omitted in Husain's tr., 70). Venkataramanyya, Early Muslim expansion, 120-1. On

Gutti (Gooty) town, at 15° 7' N., 77° 39' E., see IG, XII, 327-9.

66 FS, 427-30 (tr. 654-8). IB, III, 318-20 (tr. Gibb, 710-11). TFS1, Bodleian ms., fol. 192b/ Digby

Coll. ms., fol. 161a.

~provinces ruled by Muhammad;67 although within a few years it became part of the kingdom of Vijayanagara (c. 1336).

About the extension of the sultan's influence to the coast of Maharashtra, little evidence is

available. A European traveller tells us that Tana had been forcibly incorporated into the Delhi Sultanate by

c. 1321,68 but Ibn Battuta suggests that the Hindu ruler of the uplands between Dawlatabad and the Konkan

('Kukan Tana', as he calls it) was independent at the time of Hushang's revolt.69 Judging by the same

author's testimony, the rulers of the Malabar coast were independent, with the exception of the Muslim

prince of Hinawr, who was then subordinate to the rising power of Vijayanagara.70 If Ibn Battuta's claim

that the Muslim rulers of the Maldives feared the Delhi Sultan, despite the distance that separated them

from his dominions,71 is well grounded, it must have been a fortiori more true of the rulers of Malabar,

whose territories lay on the fringes of the Delhi empire.

The extensive territories south and south-east of the Yadava dominions had begun to attract the

attention of ‘Ala' al-Din and his officers as early as c. 701/1301-2, when the sultan's brother Ulugh Khan

had died while gathering troops at Ranthanbor for an expedition to Tilang and Ma'bar.72 Doubts have been

expressed concerning the final destination of levies from Awadh and Kara which ‘Ala" al-Din despatched

in 702/1302-3 to attack Tilang; but epigraphical evidence reveals an engagement with the Muslims near

Upparapalli (in the present-day Hyderabad State) not long before 1304.73 Barani tells us simply that the

troops became bogged down in the monsoon rains and the campaign was a failure.74 Then in 709/1309

Kafur was sent to Tilang. The Delhi forces invested Arangal (Warangal), capital of the Kakatiya king

Rudradeva II (the 'Laddar Deo' of Muslim authors), and had taken the outer, clay walls of the fortress when

Rudradeva asked for terms. He was left in peace in return for a written agreement to provide an annual

tribute.75 In 711/1311-12 Rudradeva duly forwarded a number 67 TFS, 467.

68 Odoric of Pordenone, 'Relatio', vii, 5, in Van den Wyngaert, Sinica Franciscana, I, 423, and tr.

in Sir Henry Yule, Cathay and the way thither, 2nd edn by H. Cordier, HS, 2nd series, 33, 37, 38, 41

(London, 1913-16,4 vols.), II, 114-15.

69 IB, III, 335-6 (tr. Gibb, 718).

70 Ibid., IV, 67-8 (tr. Gibb and Beckingham, 803-4): he calls the Vijayanagara king (Harihara)

'Haryab'. 71 Ibid., IV, 158 (tr. Gibb and Beckingham, 843). 72 TFS, 283.

73

Venkataramanyya, Early Muslim expansion, 24-5.

74 TFS, 300; slightly fuller in TFS1, Digby Coll. ms., fol. 113. K. S. Lal, 'A note on Alauddin's

expedition to Warangal (1302-3 A.D.)', JUPHS 16, part 1 (1943), 118-24, and History of the Khaljis, 78-

80, develops an unconvincing line of argument that this expedition was actually sent against Bengal rather

than to Tilang by way of Bengal and Orissa, as Barani claims.

75 TFS, 329-30; see 326-7 for the date. Wassaf, 527, briefly refers to this campaign, which he says

was led by 'Malik Nabu', Zafar Khan and 'Nanak [the printed text reads BABK] the Hindi': the latter two

commanders are not mentioned in other accounts of the expedition.

~of elephants to Delhi as a gesture of submission.76 After ‘Ala' al-Din's death, however, he

evidently forgot his promises, for in 718/1318, towards the end of Qutb al-Din's Deogir campaign,

Khusraw Khan was sent to extort tribute from Arangal. Once again Rudradeva yielded before the Delhi

troops could breach the inner fort; once again he handed over treasure and elephants and entered into

undertakings for the future, receiving in exchange a chatr, a durbash and a jewelled robe. Khusraw Khan

had initially demanded the surrender of five districts, Bidar (Bi-darkot), Kailas, Bodhan, Alur and Koyir

(Koher), but at length agreed to be content with Bidar.77 Yet the overthrow of the Khaljis and the events

leading to the accession of Ghiyath al-Din Tughluq in 720/1320 evidently enabled Rudradeva to repudiate the overlordship of Delhi a third time and to reoccupy Bidar.78 The new regime seems to have decided on

his removal. In 721/1321-2 an army led by the sultan's son Ulugh Khan (the future sultan Muhammad)

invested Arangal. These operations were abandoned owing to a mutiny on the part of some leading amirs

(above, p. 180), but on the arrival of reinforcements from Delhi the prince returned to Tilang, taking Bidar

and threatening Bodhan, whose rai yielded and accepted Islam. Then he again invested Arangal, rejecting

Rudradeva's offer to resume payments of tribute. Arangal fell after a five-month investment, and

Rudradeva was sent off to Delhi, only to die en route. Ulugh Khan, who remained in the south for some

time, brought Tilang under subjection, appointing governors, muqtacs and revenue officers for the new

province and taking one year's land-tax (khardj). Arangal itself, which is found as a mint-town a few years

later, was renamed Sultanpur.79

The sultan's armies penetrated the eastern coastal regions only rarely. Khusraw Khan raided

Motupalli (Marco Polo's 'Mutfili') on his way from Tilang to Ma’bar in 718/1318;80 and in the wake of his

second Tilang expedition Ulugh Khan invaded Jajnagar, routing the king's army and gaining a considerable

plunder.81 Al-cUmari was told that he had conquered the country, and lists Jajnagar among the provinces of

the Sultanate.82 But

76 TFS, 334.

77 NS, 114-35, gives the most detailed account: see 128, 132 for the territorial stipulations; the

printed text reads BDRKWB, BSWDN and KWBR. FS, 361-3 (tr. 560-2). Venkataramanyya, Early Muslim

expansion, 83-6, discusses the conflicting testimony in NS and FS regarding this campaign.

78 Ibid., 97-8.

79 TFS, 446-50. TMS, 95. FS, 392-6, 400-2 (tr. 597-603, 606-9; some lines omitted at 608), alone

mentions Bodhan. For Rudradeva's death, see ‘Afif, 395; also Hodivala, Studies, I, 337-8, and

Venkataramanyya, Early Muslim expansion, 119-20 and n.38. For coins from 'Sultanpur' (729/1328-9

onwards), see CMSD, 118 (no. 482), 120 (no. 486), 142 (no. 593A).

80 TMS, 85. For this territory, see Marco Polo, tr. Moule and Pelliot, I, 394-7/tr. Yule and Cordier,

II, 359-61 and n.l at 362; Pelliot, Notes on Marco Polo, 787-8. Motupalli lies at 15° 43'N., 80° 20'E.

81 FS, 402-3 (tr. 609-11). TFS, 450, is laconic. 82 MA, ed. Spies, 5, 6 (German tr. 24, 26)/ed. Fariq, 11, 14 (tr. Siddiqi and Ahmad, 29, 30).

~an inscription of 724/1324 from Rajahmundri, in the Godaviri delta and doubtless close to the

Jajnagar kingdom's southern frontier, may well be the only memorial of Ulugh Khan's 'conquests' here.83

The relationship was purely a tributary one. When Sultan Firuz Shah invaded Jajnagar some decades later,

the rai, Virabhanudeva III of the Eastern Ganga dynasty, claimed that he and his father had both been

servants of the court of Delhi.84 But ‘Afif, whose father had accompanied the sultan, observes that the

country contained no Muslims.85

The far south

The wealth of the Coromandel coast, known to the Muslims as Ma’bar, was proverbial, and had

attracted comment from Marco Polo at the turn of the century.86 In 710/1310-11 Malik Na'ib Kafur

advanced on Ma'bar by way of Dvarasamudra, whose Hoysala king, Ballala III, was just about to exploit

the civil war in Ma'bar (below): taken by surprise, he submitted and acted as guide to the sultan's forces.87

This subservience persisted, for when Kafur withdrew north in the wake of the Ma'bar campaign he took

with him to Delhi Ballala Ill's son Vira Ballala, who did obeisance to 'Ala' al-Din and was rewarded with a

robe (khil'at), chatr and treasure before being sent back with honour to Dvarasamudra.88 Thereafter we

know little of Ballala Ill's activities. Although he seems to have asserted his autonomy after the fall of the

Khalji dynasty, he was not disposed to defy Muslim armies. When Garshasp took refuge with him from

Kampila in c. 1327, he made no attempt to emulate the raja of Kampila but duly handed over the fugitive to the representatives of Muhammad b. Tughluq.89

Kafur, who reached the borders of Ma'bar in Shawwal 710/March 1311, was less successful here

than in Dvarasamudra, despite the opportunities offered by the civil war within the kingdom. According to

rumours that reached Persia, the king had been murdered in 709/1309-10 by his son Sundara Pandya, who

resented being supplanted in the succession by an illegitimate brother Vira Pandya, and a struggle then

ensued between the brothers.90 At Kafur's approach Vira Pandya fled from his capital at Viradhavelan

(Amir Khusraw's 'BirdhuT), and Kafur abandoned the search

83 G. H. Yazdani, 'Inscription of Ghiyathu'd Din Tughluq from Rajahmundry', EIM (1923-4), 13-

14. 84 SFS, 67 (tr. Roy, 'Jajnagar expedition', 72); and see also 'Afif, 171.

85 Ibid., 165; for 'Afif's father, 163.

86 Marco Polo, tr. Moule and Pelliot, I, 381-6/tr. Yule and Cordier, II, 338-40.

87 KF, 127; date of Kafur's departure ibid., 116. FS, 293-5, 297 (tr. 468-70, 471).

88 Ibid., 298 (tr. 473). Venkataramanyya, Early Muslim expansion, 67 and n.129.

89 FS, 431 (tr. 658-9). IB, III, 321 (tr. Gibb, 711), does not mention Ballala by name. J. Duncan M.

Derrett, The Hoysalas: a medieval Indian royal family (Oxford, 1957), 162-4. Venkataramanyya, Early

Muslim expansion, 143-4.

90 Wassaf, 530-1. KF, 127, briefly refers to the parricide and the conflict. For the date of Kafur's

arrival, see ibid., 143: five days after his departure from Dvarasamudra (ibid., 142).

~for him when it became clear that the king had taken refuge in the jungle. Sundara Pandya in turn

abandoned his residence at Mathura (Madura) prior to the arrival of the sultan's army. But in Dhu'l-Hijja

710/April 1311 Kafur withdrew from the country.91 The Delhi forces had been impeded by the monsoon

rains, and reports reached Persia that a large army had been mustered against them.92 In the wake of Kafur's

attack, the brothers continued their conflict, in which Ma'bar's neighbours, the sultans included, were only too happy to intervene. Sundara Pandya was defeated and took refuge with 'Ala' al-Din's forces

(presumably at Deogir), with whose help he had, by the beginning of 1314, re-established himself in the

South Arcot district.93 In c. 718/1318 Qutb al-Din Mubarak Shah, fresh from the suppression of a rebellious

Muslim governor at Deogir, sent Khusraw Khan against Ma'bar; the city of Pattan was taken and sacked,

and the Delhi forces acquired an enormous plunder.94

The real advance of the sultan's armies in this region, however, dates from the reign of Ghiyath al-

DIn Tughluq. Muslim sources tell us nothing of the conquest, although Sirhindi asserts that Ulugh Khan

was sent against Ma'bar as well as Tilang in 721/1321. According to a Pandyan chronicle, however, the

reduction of Ma'bar, along with the capture of a king called Parakramadeva, occurred in the Saka year 1246 (1323);95 although the temple at Srirangam may not have been destroyed until 1327.96 King Sundara

Pandya and other members of his dynasty seem still to have been acknowledged in parts of the kingdom in

the 1330s and even later, and it appears that the southernmost dominions of the Pandyas were never

absorbed into either the province of Ma'bar or the independent sultanate that replaced it after 1334.97

91 Ibid., 148, 150, 152-3, for Vira Pandya's flight; 154-5 for the abandonment of the search; 160

for Sundara Pandya's flight from Madura; 166 for Kafur's withdrawal. 'Birdhul' (Uyyakkonddan Tirumalai,

a few miles from Uraiyur) is identified by V. Venkatasubha Aiyar, 'Srirangam inscription of Kakatiya

Prataparudra: Saka 1239', EI 21 (1947-8), 311; Derrett, Hoysalas, 233. For the failure of the campaign, see

Venkataramanyya, Early Muslim expansion, 65-7.

92 KF, 150-1. Wassaf, 528. 93 Ibid., 531: Wassaf, the sole Muslim source to mention Sundara Pandya's appeal to ‘Ala' al-Din's

forces, gives the false impression that it occurred during Kafur's invasion of 1311. Venkataramanyya, Early

Muslim expansion, 88-90 and n.16.

94 FS, 369-71 (tr. 569-72). See Venkataramanyya, Early Muslim expansion, 93-4, for these

operations: as he points out, TMS, 84-5, links up the two quite separate campaigns against Tilang and

Ma’bar.

95 TMS, 93. Venkataramanyya, Early Muslim expansion, 122-5; see also ibid., 70 and n.136, for

the date. HN, 472. 96 G.W. Spencer, 'Crisis of authority in a Hindu temple under the impact of Islam: Srirangam in

the fourteenth century', in Bardwell L. Smith (ed.), Religion and the legitimation of power in South Asia

(Leiden, 1978), 20-3 and n.18.

97 Venkataramanyya, Early Muslim expansion, 156 n.15. ARIE (1980-1), 5, 77 (no. B199). K. G.

Krishnan, 'New light on Madurai Sultanate', in PSMI, 156-7.

~War aims and achievements

The initial purpose of the campaigns into peninsular India was to obtain plunder and the guarantee

of tribute. In the advice to ‘Ala' al-Din which Barani puts into the mouth of his own uncle ‘Ala' al-Mulk, the sultan is urged to leave in the hands of the rais and ranas no elephants, horses or wealth and to require

these things every year.98 Vanquished Hindu rulers were regularly mulcted of their treasure. The enormous

tribute which Khusraw Khan imposed on Rudradeva of Tilang, even when reduced, stood at 48 laks

(4,800,000) of gold coins." Temples, too, yielded up large quantities of gold, like that at Birdhul or the

golden temple at the place called both 'Barmatpuri' and 'Marhatpuri by Amir Khusraw, which Kafur left in

ruins during his Ma'bar campaign.100 Plunder on such a scale rapidly acquired a legendary character. Barani

claims that in his own day some of the riches disgorged by Ramadeva in 695/1296 were still to be found in

Muhammad b. Tughluq's treasury, while the amount obtained from Dvarasamudra and Ma'bar in 710/1311

was indeed phenomenal, allegedly totalling 96,000 manns of gold as well as gems and pearls - a booty that

evidently made a profound impression on Delhi's older residents.101

In Ma'bar Hindu princes and temples were not the only victims of predatory Muslim commanders.

During the 718/1318 campaign Khusraw Khan is accused of despoiling a wealthy and respectable Muslim

merchant who had not judged it necessary to flee before an army led by his coreligionists. This person,

whom Barani calls Khwaja Taqi and who appears in Tsami's more detailed account as Siraj-i Taqi, charge

d'affaires (farman-nuwa) in Pattan,102 belonged to the dynasty which controlled the island of Qays in the

Persian Gulf. According to Wassaf, his uncle Taqi al-DIn cAbd al-Rahman (d. 702/1302-3), wazir and

counsellor to the king of Ma’bar, had been responsible for the importation of war-horses from Qays and

adjacent regions.103 Wassaf also tells us that Siraj al-Din's property had been looted during an invasion of

Ma'bar by Kafur in 715/1315 (possibly the one in support of Sundara Pandya: see p. 207), just before the

death of ‘Ala' al-Din Khalji, but that it had been restored to him when he complained. Since Siraj al-Din's father is here said to have enjoyed friendly relations with ‘Ala' al-Din,104 it is possible that Khusraw Khan's

actions a few years later reflect a change in policy; but no doubt the conqueror was simply greedy for Siraj

al-Din's wealth.

98 TFS, 270. 99 NS, 128, 132. 10° KF, 156-9, 160. DR, 72.

101 TFS, 223, 333-4. Hodivala, Studies, II, 103-5, discusses the large quantities of gold obtained

in 1311. It is unlikely that we can base our calculations, as he did, on the Delhi mann of the thirteenth-

fourteenth centuries, which ranges from 11.25 to 12.824 kg.: see Walther Hinz, Islamische Masse und

Gewichte (Leiden, 1955), 22-3.

102 TFS, 398-9. FS, 369-70 (tr. 570-1). 103 Wassaf, 302-3, 505. On the dynasty, see Aubin, 'Les princes d'Ormuz', 89-99.

104 Wassaf, 646-7.

~Elephants, horses and specie loom large as both plunder and tribute in the chroniclers' accounts.

Precise figures for the horses obtained on these southern campaigns are sometimes given in the sources.

'Ala' al-Din had obtained several thousand on his Deogir campaign of 695/1296.105 Kafur brought back

20,000 horses from Arangal in 709/1310 and 5000 Yamani horses from Ma'bar two years later.106

Rudradeva handed over 12,000 Arabian (tazi) horses to Khusraw Khan and promised to send 1000 every

year in future.107 In comparison, the figures for elephants sometimes seem rather modest: from Gujarat in

698/1299, 20; from Ramadeva, thirty or so in 695/1296 and a further seventeen in 706/1307; forty from Jajnagar in 1324.108 From Rudradeva of Tilang in 709/1309 Kafur extorted a hundred, while after he had

crossed the Narbada during his Macbar expedition in 710/ 1311 the king sent him another twenty-three,

which Kafur forwarded to 'Ala' al-Din at Delhi.109 In the course of Khusraw Khan's attack, Rudradeva

offered 100 elephants, and the victor stipulated that 100 should be sent annually.110 The acquisition of large

numbers of high-quality elephants appears to have been a major aim of the invasion of Ma'bar in 710/1311.

It is mentioned as such in Khusraw's Diwal Rani and by 'Isami in his account of the sultan's instructions to

Kafur. Certainly Kafur came to give priority to the seizure of elephants even over the capture of Vira

Pandya, and Khusraw describes his fury at finding only two or three of the beasts in Madura.111 Yet despite

such disappointments the Ma'bar campaigns yielded significantly larger numbers of elephants than did raids

further north. Amir Khusraw says that Kafur brought back to Delhi 512 of them, although this may have

made things more difficult for those who came after him, since Khusraw Khan in 718/1318 captured hardly

more than a hundred.112

The transition from a policy of plunder and levying tribute to one of imposing direct rule, already

made in the Deccan a few years previously, is visible during the attack on Arangal in 721/1321, when

Ulugh Khan rejected Rudradeva's offer of submission and pressed ahead with the siege.113 One may well

ask, nevertheless, to what extent these far-flung provinces were

105 TFS 223.

106 KF, 101, 163. TFS, 330, numbers 7000 horses among the booty from Arangal in 709/1310;

ibid., 333, for 20,000 from Ma’bar in 710/1311.

107 NS, 120, 128, 132.

108 Gujarat: TMS, 76. Deogir: TFS, 223, 326. Jajnagar: ibid., 450; hence TMS, 96.

109 DR, 70; TFS, 330. FS, 291 (tr. 466), specifies twenty-three on the former occasion, but these

are clearly the elephants despatched in 1311: KF, 120; Venkataramanyya, Early Muslim expansion, 39-40

and n.56. TFS, 334, mentions their arrival at Delhi (though giving the total as twenty).

110

NS, 120, 128, 132. See also FS, 362 (tr. 561), for the initial surrender of 100 elephants.

111 DR, 70. FS, 293-4 (tr. 468). KF, 155, 160.

112 1311: ibid., 161: the total of 612, of which thirty-six were taken from Dvarasamudra, found in

TFS, 333, is probably an error; so too is the round figure of 700 in FS, 298 (tr. 472). 1318: TFS, 398, 400.

113 TFS1, Bodleian ms., fol. 183b/Digby Coll. ms., fol. 155a. TFS, 447.

~ever truly annexed. Vast distances separated Delhi from its new provinces: Ibn Battuta believed,

with pardonable exaggeration, that Tilang was three months' journey from the capital and Macbar six.114

Such distances gave rise to the most alarming delays in the transmission of news. The fourteenth-century

sultans extended the postal relay system to the outlying regions of their empire.115 But sometimes it broke

down; armies receded beyond the horizon of communications and appeared to have been swallowed up in some limbo zone. In 721/1321-2 the commanders outside Arangal mutinied because a delay of a few

weeks, in which couriers failed to get through from Delhi, spawned rumours that Sultan Tughluq had been

overthrown.116 There is sometimes a starkly unreal quality about the links that bound such remote territories

to their imperial master.

The Deccan recognized the Delhi Sultan for less than thirty years; eastern Tilang, Kampila and

Macbar, for an even briefer interval. And yet the transient rule of the sultans bequeathed to the Deccan one

legacy, of major importance. Because this region had a strategic value relative to the other southern

kingdoms, the Khalji and Tughluqid monarchs made positive efforts to bring about Muslim colonization of

the former Yadava dominions. As a consequence, this territory alone - when barely more than a generation,

astonishingly enough, had elapsed since its conquest - had received a solid engrafting of Islam. The other southern provinces swiftly repudiated the sultans' faith along with their sovereignty and reverted to the

infidel. But the impress of some years' subjection to Delhi would remain, even so, in the culture and

titulature of the Vijayanagara court, where the fourteenth-century monarchs styled themselves 'sultans

among Hindu kings'.117

Striking testimony to the government's authority in the Deccan emerges from an incident during

the mutiny in Tilang in 721/1321-2, when Mujir al-Din Abu Rija, the mushrif of Deogir, met the mutineers

at Kalyani at the head of a large number of landholders (zamindars) - presumably the Hindu landed

gentry.118 Muhammad b. Tughluq's own efforts from 727/1326-7 to turn Deogir, now renamed Dawlatabad,

into the second capital of his empire (below, pp. 258-60), could perhaps be seen, at one level, as the most

impressive witness to the strength of Muslim rule here. But even in the Deccan, where it became firmly

established, Muslim rule was uneven and extended to only a limited number of strongpoints by the time the province seceded from Delhi.

The consolidation of Muslim rule and implantation of Islam are processes largely hidden from us.

A significant role may have been played by warrior

114 IB, III, 192, 208, 328 (tr. Gibb, 644, 652, 715). 115 E.g., TFS, 330-1.

116 Ibid., 447-8.

117 Phillip B. Waggoner, '"Sultan among Hindu kings": dress, titles, the Islamicization of Hindu

culture at Vijayanagara', Journal of Asian Studies 55 (1996), 851-80. 118 FS, 398-9 (tr. 604-5).

~Sufis whose activities are described in later hagiographical sources. An account has survived of

the career of one such warrior saint, Ma’bari khandayat, in Bijapur in the wake of Kafur's Ma’bar campaign

of 710/ 1311, and Professor Eaton has made out a good case for accepting the outline as authentic and for

identifying Khandayat as one of the Muslims formerly in the service of the Pandyas.119 The militant sufi

‘Abd-Allah Shah Changal, to judge from the mid fifteenth-century inscription on his tomb, seems to have

entered Malwa at the head of a military following and played much the same role in the conversion of that

province; his activities too are in all probability to be assigned to the era of ‘Ala’al-Din.120 But it should be borne in mind that not all sufis resorted to force, for Ibn Battuta learned that the infidels of Sylhet had been

won for Islam by the peaceful agency of Shah Jalal.121

We seldom hear of specific territories being granted as iqta’s during the years immediately

following the conquest, although Sagar, south of Gulbarga, was conferred before 1326 on Baha' al-Din

Garshasp, a cousin of Muhammad b. Tughluq;122 and the gradual build-up of Muslim authority in the

Deccan can be determined only to a limited extent on the basis of epigraphical evidence. Inscriptions show,

for instance, that Jalna, only a few miles from Deogir, was under Muslim occupation by 724/1324 and that

Bhadga'un (in eastern Khandesh) received a mosque in 728/1328;123 Bijapur was already the seat of a

Muslim governor by 1320, when a mosque was built in the town.124 Otherwise, to form some idea of the

number of centres under Muslim control we must rely on 'Isami's account of events in the Deccan in the

1340s. There we learn, for example, that Dangiri and Chanchiwal, in the north-west of the former Yadava realm, had to be taken from Hindu chiefs by Bahmanid troops in c. 1350.125 But the strongpoints of which

we hear tend mostly to be concentrated in the south and south-east of the province, in an arc between

Deogir and the erstwhile Kakatiya capital of Arangal. Here Gulbarga, Bidar, KalyanI and Koyir (Koher)

appear as a compact group of Muslim-held fortresses.126 Bidar - like Bodhan to the north - had been taken

from Rudradeva by Ulugh Khan in c. 1322, and in KalyanI we find mosques being constructed in the

1320s.127 On the other hand,

119 Richard M. Eaton, Sufis of Bijapur 1300-1700 (Princeton, NJ, 1978), 27-30.

120 G. H. Yazdani, 'The inscription on the tomb of ‘Abdullah Shah Changal at Dhar', EIM (1909-

10), 1-5; and see ARIE (1971-2), 80 (no. D71). More generally, see David N. Lorenzen, 'Warrior ascetics in medieval Indian history', JAOS 98 (1978), 61-75.

121 IB, IV, 216-17 (tr. Gibb and Beckingham, 870). Eaton, Rise of Islam, 73-6.

122 FS, 424-5 (tr. 651).

123 Jalna: ARIE (1964-5), 23, 153 (no. D161). Bhadga'un: Moneer, 23-4.

124 M. Nazim, Inscriptions of Bijapur, MASI 49 (Delhi, 1936), 25.

125 FS, 560 (tr. 834).

126 Gulbarga: ibid., 485 (tr. 726). Bidar and Koyir: ibid., 476 (tr. 715).

127 723/1323: ARIE (1965-6), 155 (no. D246), correcting the earlier reading of G. Yazdani

~Map 4: The conquest of Gujarat, Malwa and the south

the neighbouring fortresses of Maram, Akalkot and Mahandari (Ma-hendri) were still in the

possession of infidel rais at the onset of the Bahmanid era.128 Towards the Western Ghats, Satara and Miraj

were in Muslim hands by the 1340s.129 But just south of Miraj, Balga'un (now Belgaum) and Hukayri

(Hakeri), at that time the iqtac of the future Bahmanid Sultan, Hasan Gangu, are described repeatedly by

'Isami as a marcher lordship (sarhadd).130 Close by these tracts lay Mandhol (present-day Mudhol), Jamkhandi, Terdol and Bagarkot (now Bagalkot), the territories of the independent Hindu prince Narayan,

who would prove a redoubtable antagonist for the infant Bahmanid regime.131

Barani says that in 709/1309 the chiefs (muqaddams) of Tilang abandoned the strongholds along

the route taken by the sultan's army; whether

'Inscriptions from Kalyani', EIM (1935-6), 1-3. 726/1326: ARIE (1965-6), 14-15, 157 (no. D271).

128

FS, 562 (tr. 836). See H. K. Sherwani, The Bahmanis of the Deccan (Hyderabad, AP, 1953),

53. 129 Satara: FS, 519-20 (tr. 770). Miraj: ibid., 540-2 (tr. 811-12).

130 Ibid., 521, 526, 532 (tr. 772, 778, 785). 131 Ibid., 590-1 (tr. 871-2).

~garrisons were immediately installed in such places we are not told.132 But some fortresses put up

resistance. Kafur had to halt at Sirpur, which he took and entrusted to the brother of its chief.133 Kotgir,

which Mujir al-Din Abu Rija was besieging several years later, at the time of the mutiny against Ulugh

Khan, seems nonetheless to have remained in enemy hands, for in the early 1340s Qutlugh Khan, who

governed the Deccan on behalf of Muhammad b. Tughluq, took Kotgir from a Hindu 'rebel' and stationed

there one of his own lieutenants. The Chandagarh (Chanda, i.e. Chan-drapur) region, which he sent his son

to plunder at around this time, was clearly independent under its own Hindu princes.134

Muslim military superiority

'The Hindu always falls prey to the Turk', wrote Amir Khusraw in his Nuh SipihrP5 A little later,

having likened the Turk to the lion and the Hindu to the gazelle, he claims that the Turks, whenever they

bestir themselves, can vanquish the Hindus and may seize and buy and sell them.136 'A mere six or seven

thousand Muslim horsemen,' Barani makes Sultan Balaban tell his sons, 'could rout one lak of Hindu paiks

and archers (dhdnuks).'ul Marco Polo, commenting that the men of Ma'bar - whose only defence in battle

was shield and spear - made wretched warriors, was doubtless citing Muslim informants.138 The superiority

of the Muslim troops is almost a commonplace in our (Muslim) sources. Satisfactory explanations for it are

less forthcoming. There is clearly a link between the assertion of Muslim paramountcy throughout the greater part of the subcontinent and ‘Ala' al-Din's administrative reforms, which enabled the sultan to raise

larger numbers of troops on lower pay and which will be examined in a later chapter.139 At times ‘Ala' al-

DIn's troops also profited from the fact that their Hindu adversaries were bitterly divided, as in Malwa in

705/1305 or -at least after their first unsuccessful attempt - in Ma’bar.

Observers within the Sultanate, however, thought they could account for the sultans' victories over

the Hindu on technical grounds also. The Hindus were not good marksmen, according to Khusraw;140 there

was no force in

132 TFS, 329. 133 KF, 80-2; for the identification, see p. 202, n.54 above.

134 FS, 397-8, 482-3, 500-1 (tr. 603-4, 723-5, 747-8). The identification of the last seems fairly

certain, since the Muslim army is said to have gone by way of Akola ('Ankula' in the text).

135 NS, 89, Hindu buwad sayd-i Turkan hamisha. Cf. the view attributed to Ballala III in KF, 131,

hargaz Hindu plsh-i Turk ... tab nayarad.

136 NS, 130, 131. 137 TFS, 52.

138 Marco Polo, tr. Moule and Pelliot, I, 389/tr. Yule and Cordier, II, 342. For other allusions to

Muslim superiority in the sources, see Aziz Ahmad, 'Epic and counter-epic in medieval India', JAOS 83

(1963), 470-1.

139 TFS, 3O3ff., 326.

140 KF, 135 Hindiianra klsh-i durust riist (speech put into the mouths of Ballala Ill's envoys). Cf.

also the description of the envoys themselves, ibid., 137, kaman-war-i kazh-nishin.

~their arrows, remarked Ibn Battuta.141 It has been proposed that the sultans' Indian opponents

never adopted mounted archery.142 This may perhaps have been true of certain Hindu armies. ‘Isami, for

instance, characterizes the troops of Deogir led by Ramadeva's son in 695/1296 and those of Jajnagar

whom Ulugh Khan defeated in c. 1322 as 'all spear-wielders and swordsmen', though the phrase could owe

more to style than to critical observation.143 The Delhi forces conceivably enjoyed a superiority in terms of certain types of weaponry. ‘Is.ami's account of Kafur's Arangal campaign gives some prominence to the

crossbow (ndwak; see p. 16 above), and when Ulugh Khan attacked Arangal for the second time he is said

to have taken both the outer and inner defences by dint of firing ndwaks and stones from catapults.144 The

nawak certainly figures prominently in the catalogue of weaponry employed by the sultan's troops.145 An

anecdote of 'Isami's, in which a guruha fired by a 'Turk' not only penetrated the wheel of a wagon but

embedded itself in the earth beyond right up to the feather, suggests that this particular weapon (expressly

called an arrow, and hence presumably a crossbow-bolt) was calculated to strike terror into the enemy.146

It is fair to say, on the evidence of the narrative sources, that the sultan's forces were seldom

granted the opportunity of a pitched battle. On the few occasions when it did happen, the Delhi army is

portrayed as having won an almost effortless victory - as when the army of Malwa challenged 'Ayn al-Mulk

in 705/1305, or when Ramadeva's son marched out to challenge Kafur in 706/1307 or when Vira Pandya's rawats met Kafur outside Birdhul four years later.147 But Rudradeva is expressly said to have avoided a

pitched battle in 721/1321-2.148 The reason for this apparent Muslim advantage may have been a chronic

inability on the part of many Hindu rulers to match the sultan's armies in terms of horses; and the eagerness

of the monarchs of peninsular India to obtain horses in large numbers from Arabia and the Persian Gulf

was notorious.149 There are already indications in the middle of the thirteenth century of an imbalance in

this respect

141 IB, III, 134 (tr. Gibb, 613).

142 P. K. Gode, 'The mounted bowman on Indian battle-fields - from the invasion of Alexander

(B.C. 326) to the battle of Panipat (A.D. 1761)', in his Studies in Indian cultural history (Hoshiarpur and Poona, 1960-9, 3 vols.), II, 57-70. Wink, Al-Hind, II, 82-3. For mounted combat, see Digby, War-horse, 12

and n.5.

143 FS, 234, 402 (tr. 403, 609).

144 Kafur: ibid., 290-1 (Husain's tr., 465, does not bring out the sense). Ulugh Khan: TFS, 449;

TMS, 95.

145 E.g. KF, 55, 56, 57, 58-9, 80-1, 93, 128, 136, 150. Perhaps this is the arrow that pierces seven

plates of iron: ibid., 96.

146 FS, 230 (tr. 397-8); see ibid., 54 (tr. 108), for a gur/ta [sic] which transfixed a deer during one

of Mahmud of Ghazna's campaigns. By contrast, the guriiha-yi maghribl mentioned in KF, 90, was clearly

fired from a mangonel.

147 KF, 65-6, for Deogir; 151-2 for Birdhul. !48 TFS, 446.

149 Digby, War-horse, 29-32. Wink, Al-Hind, II, 83-7.

~between the Muslims and Hindu rulers, as when Juzjani alleges that the Jajapella king

Chahadadeva possessed a mere 5000 horse as against 200,000 foot or when he mentions only paiks and

elephants in the army that the king of Orissa brought into Muslim Bengal in 642/1244.150 Similarly, at the

very end of the century Hammlradeva of Ranthanbor is credited with 'countless infantry' but just 12,000 horse.151 Yet, while such figures suggest that the Delhi forces enjoyed a greater striking power than their

Hindu opponents, in other cases the proportion of cavalry to infantry would seem to have been roughly

similar on both sides. Karnadeva of Gujarat, for instance, had 30,000 horse as against 80,000 foot in his

army, and Koka in Malwa had 40,000 horse and one lak of foot.152

Whatever the case, the leitmotiv of the Khalji and Tughluqid campaigns both north and south of

the Vindhyas is one of sieges. An inscription of 1261 at Ajayagarh calls the Chandella king

Trailokyavarman 'a very creator in providing strong places', and it has been suggested that this provides a

genuine hint as to the tactics followed in resisting Muslim incursions.153

It may be that Muslim siege

warfare of the early fourteenth century represents an advance on that of the Shamsid and Ghiyathid eras, but unfortunately neither the thirteenth-century sources nor those for 'Ala' al-Din's reign furnish enough

information to warrant firm conclusions, and Professor Lal's case that the Muslims enjoyed a definite

superiority in this respect must be regarded as unproven.154 Amir Khusraw says that the walls of

Ranthanbor were demolished by maghribis (mangonels); but we know, on the other hand, that Chitor

surrendered and that Mandu was taken through the treachery of a Hindu deserter, who showed 'Ayn al-

Mulk a way into the fortress. At the investment of Siwana the sultan's troops constructed a pdshib, a

gradated platform made out of earth, mounting to the level of the walls, and this was clearly important in

their success.155 So too the pashib raised by Khusraw Khan for the investment of the inner fortress at

Arangal in 718/1318 was instrumental in bringing Rudradeva to ask for terms.156 At other times, however,

the role of the pdshib is difficult to assess, since it is also clear that such a device had earlier been employed

at Ranthanbor and had suffered considerable damage from the enemy cata-

150 Jajapellas: TN, I, 485 (tr. 691): for this and other relevant evidence, see Digby, War-horse, 49.

Orissa-. TN, II, 15 (tr. 739).

151 TMS, 77. 152 The figures are from ibid., 76.

153 P. Prasad, Sanskrit inscriptions, 100-5 (verse 7). Ray, Dynastic history, 727.

154 K. S. Lal, 'The striking power of the army of the Sultanate', JIH 55 (1977), part 3, 100-1.

155 Ranthanbor: DR, 65-6. Chitor: KF, 62; FS, 281 (tr. 456). Mandu: KF, 58; DR, 68, mentions a

'breach'. Siwana: KF, 50-1, 70; TMS, 78, indicates that the fortress was stormed. Hodivala, Studies, I, 112, renders pashlb, somewhat loosely, as 'earthworks'; for this and other devices, see Athar Ali, 'Siegecraft

techniques of the Delhi Sultans during 13th and 14th century [sic] PIHC 51. Calcutta 1990 (Delhi, 1991),

217-26, and his 'Military technology', 171-3.

156 NS, 111-14.

~pults.157 At Arangal in 709/1309 the besiegers did not wait for the completion of the pashib

before launching their assault on the outer, mud wall.158

In the current state of our knowledge, questions about the capacity of the sultans' armies to

vanquish their Hindu opponents are unanswerable. But if the inferiority of such antagonists was more often

than not taken for granted, developments elsewhere had put it in a new perspective. 'Where should the army that defeats the Mongol host be afraid of fighting the Hindu?' asked 'Isami sardonically.159 Long before the

time at which he wrote, the sultans had been given greater reason to hate and fear the other infidel enemy,

to the north-west.

157 KF, 50-1. TFS, 277. 158 KF, 91. 159 FS, 284 (cf. tr. 459).

~CHAPTER 11

The Chaghadayid invasions

The southward advance of the Central Asian Mongols

Chapter 6 surveyed the disintegration of the unitary Mongol empire, culminating in the creation of

a confederacy of princes under Ogodei's grandson Qaidu in Central Asia in opposition to the qaghan

Qubilai. Although Qaidu's own campaigns seem always to have been directed against Qubilai's lieutenants

and supporters in Mongolia, he also pursued, if less directly, an expansionist policy south of the Oxus. The

rulers of Chaghadai's ulus, whom he nominated, appear to have acted as his subordinates: the last and most

important of them was Du'a, Baraq's son, who became khan in c. 681/1282.1 Under Qaidu's aegis Du'a, in

Rashid al-Din's words, 'gradually gathered together the armies of Chaghadai';2 and their forces collaborated

both in eastern Persia and in Afghanistan. Qaidu's son Sarban and one of Du'a's chief noyans, Yasa'ur, were

stationed south of the Oxus by 690/1291.3 That the allies were already seeking, at this early date, to assert their influence among the Neguderis is clear from Qaidu's dealings with the renegade Ilkhanid general,

Nawruz, who operated on his behalf in Afghanistan from 690/1291 until he rejoined the Ilkhan in 694/

1294.4 Wassaf, whose account of these events is geographically more specific than Rashid al-Din's, has

Nawruz taking up his quarters in 'SIstan' (i.e. Ghur and Gharchistan), where he won over the Negiideri

forces, and says that he particularly relied on them; at another juncture Nawruz is referred to as their chief

(hakim).5 It looks very much as if Qaidu relied on Nawruz

1 Jamal al-Qarshi, 138-9. JT, II, 192-3 (tr. Boyle, 154/tr. Verkhovskii, 100). Biran, Qaidu, 32-3.

2 JT, II, 172 (tr. Boyle, 141/tr. Verkhovskii, 92).

3 Sarban: JT, ed. Karl Jahn, Geschichte Gazan Han's, GMS, ns, XIV (London, 1940), 26 (fuller

than JT, III, ed. Alizade). Yasa'ur near Balkh: JT, I, part 2, ed. I. N. Berezin, 'Sbornik letopisei', TVOIRAO

15 (1888), 217; tr. O. I. Smirnova, Sbornik letopisei, I, part 2 (Moscow and Leningrad, 1952), 275.

4 JT, ed. Jahn, Geschichte Gazan Han's, 24-6, 29, here fuller than JT, III, 268-72 (tr. Arends, 150-

2). On Nawruz, see generally Biran, Qaidu, 57-9.

5 Wassaf, 253, 314; for this sense of the term 'SIstan', see Aubin, 'L'ethnogenese', 91.

~as his agent in maintaining a fragile control over parts of Afghanistan. Towards the end of the

thirteenth century, however, that control markedly intensified.

It may have been in response to Nawruz's defection that Qaidu instituted the military dispositions

outlined by Wassaf. Sarban was put in overall command of forces totalling five tiimens (50,000), three of

them from Qaidu's own armies and two belonging to Du'a. His lieutenants included the Ogodeyid prince

Kuresbe; Temur, son of Ebugen, a descendant of one of Chinggis Khan's brothers; and Du'a's son Qutlugh

Qocha.6 Of Sarban's colleagues, the last-named, for our purposes, is the most important. According to

Rashid al-Din, Du'a recalled the Chaghadayid prince 'Abd-Allah and set Qutlugh Qocha over the Neguderis

in his place.7 The Neguderi commander Abachi, who in the early 1290s had obeyed Nawruz, now appears

as Qutlugh Qocha's subordinate.8 Rashid al-DIn says that Qutlugh Qocha spent the summer in the confines

of Ghur and Gharchistan and wintered in 'the territory of Ghaznayn and that direction'; QashanI says that he

resided in BInl-yi Gaw, which the sources, as we have seen, traditionally link with the Negiideris; while

Wassaf describes his headquarters as the valley of the 'Arghantua' (Arghandab).9 He came to rule a vast

principality, which extended from the Oxus down to the hot regions around the latitude of Qandahar.10 He struck coins at Ghazna in his own name,11 and his exalted status emerges from Rashid al-Din's allusion to

him in terms that suggest he was practically joint ruler of Chaghadai's ulus with his father Du'a, an

impression also in evidence in the Indian chronicle tradition.12 He appears to have had at his disposal

considerable reserves of manpower, since Wassaf sets his forces at five tumens; though cIsami exaggerates

in claiming that at the time of his invasion of India in 699/ 1299-1300 (below) there were 200,000 men on

his muster-roll.13

6 Wassaf, 509-10. For Kuresbe, see JT, II, 14 = II, part 1, ed. Alizade, 38-9 (tr. Boyle, 28/tr.

Verkhovskii, 17). For Temur, see SP, fols. 103b-104a; Mu'izz al-Ansab, fols. 9b, 10b, l1b: this branch of

the dynasty is omitted in JT, I, part 2, ed. Berezin, in TVOIRAO 13 (1868), 86-95 (tr. Smirnova, 51-4). Both

princes accompanied Sarban's army in 1302-3 (below): QashanI, 18. 7 JT, III, 152 (tr. Arends, 94), with the date 698/1299, which is probably too late; at II, 177 (tr.

Boyle, 144), the text is corrupt, and for the correct reading see Aubin, 'L'ethnogenese', 84 n.2.

8 Sayfi, 379-82. Wassaf, 368. KirmanI, Simt al-'Ula, 89. Aubin, 'L'ethnogenese', 88. He was

probably the unnamed Neguderi chief (shah) killed by the Ilkhan's troops in 1301: Wassaf, 417-18.

9 JT, II, 173 (tr. Boyle, 142/tr. Verkhovskii, 93). Qashani, 201. Wassaf, 367.

10 Ibid., 368, listing 'Balkh and its dependencies (madafat), Shaburghan, Juzjan, Badakhshan,

Kishm, Tayaqan, Dara-yi Suf, Dara-yi Gaz, Flruzkuh, 'Aliyabad, Malikabad, Marw (Merv) and its

appendages (lawahiq), Andkhui, Faryab, Taliqan, Maruchaq and Panjdih'. For the localization of some of

these places, see map 2; also Aubin, 'L'ethnogenese', 92 n.4.

11 Thomas, 175-6. E. Blochet, 'Les monnaies mongoles de la collection Decourdemanche', Revue

de I 'Orient Chretien 11 (1906), 119-20.

12 JT, I, part 2, ed. Berezin, in TVOIRAO 13 (1868), 125 (tr. Smirnova, 69). AHG, II, 796.

13 Wassaf, 367. FS, 256 (tr. 427).

~From these forward bases, both Sarban and Qutlugh Qocha, according to Rashid al-Din, mounted repeated attacks on the Ilkhan's eastern provinces.14 Wassaf describes the Herat region as a bone of

contention between the Ilkhan's forces and Qutlugh Qocha's Mongols.15 But the latter's raids, like those of

the Negiideris earlier, penetrated much more deeply into Persia, as when in 700/1301 (actually after

Qutlugh Qocha's death) one tumen of his forces ravaged Fars and Kirman and even rode as far as

Hurmuz.16 Both on that occasion and in 702/1302-3, when Sarban attempted to link up with Qutlugh

Qocha's troops in a joint attack on Khurasan,17 the Central Asian Mongols, who sought to profit from the

absence of the Ilkhan Ghazan in Syria, were worsted by his brother, the viceroy Kharbanda.18

At what stage responsibility for assaults on the Sultanate passed from the Negiideri bands to the

armies of Qaidu and Du'a, it is difficult to say. When Balaban's grandson Kaykhusraw sought assistance

from the Mongols at Ghazna following the enthronement of his cousin Kayqubad in 685/1287, says 'Isami, he was unsuccessful because the Mongols were preoccupied with internal disputes.19 This could

conceivably refer to the early stages of Qaidu's intervention in Afghanistan; but the evidence at our disposal

is inadequate. Even after the definite appearance of Qaidu's commanders on the scene, there were still

small-scale initiatives by what were presumably Negiideri contingents acting independently. In c.

698/1298-9 'Ala' al-Din sent his general Zafar Khan against a body of Mongols who had occupied Siwistan

in lower Sind, perhaps seeking to take advantage of the recent overthrow of Jalal al-DIn Khalji's sons and

their supporters at Multan. The invaders were dislodged, and their chief *S6gedei was captured with his

brother and brought to Delhi.20 'Isami describes him as a 'Turk' and one of his companions as a 'Baluch',21

suggesting that the episode represents a

14 JT, II, 11 (= II, part 1, ed. Alizade, 28), 173 (tr. Boyle, 25, 142/tr. Verkhovskii, 14, 93).

15 Wassaf, 368.

16 The date in JT, III, 152 (tr. 94), and supported by the lost chronicle of Hurmuz: W. F. Sinclair,

The travels of Pedro Teixeira, HS, 2nd series, 9 (London, 1902), 160-1 (but at 161 n.l, Sinclair mistakes the

marauders for the Ilkhan's forces). Fuller account in Wassaf, 368-71, with confused dating.

17 JT, II, 11-12 = JT, II, part 1, ed. Alizade, 28-30 (tr. Boyle, 25-6/tr. Verkhovskii, 14-15).

QashanI, 18.

18 Wassaf, 368. Qashani, 19.

19 FS, 196-7, Mughalra dar an waqt ba ahl-i khwlsh * magar bud digar muhimml ba-pish (tr.

329).

20 TFS, 253-4, placing the invasion in the same year as the sultan's Gujarat campaign, i.e. the

third year of the reign (ibid., 251), 697-8/1298-9; on the chronology of that expedition, see above, p. 195

and n.9. AHG, II, 787, 790, dates the occupation of Siwistan in 697 and its recovery in 698.

21 FS, 251 (tr. 421-2). The name of the invading chief, usually given as 'Saldi', appears in TFS1,

Bodleian ms., fol. 133a, as SKNY, probably an error for SKTY, i.e. Sogetei. In the critical apparatus to his

edition of FS, Usha in fact proposes SGDY. On the etymology of Sogetei/Sogedei, see Pelliot and Hambis, Histoire des campagnes de Gengis Khan, 129-30, 255-6.

~local foray from the southern parts of what is now Afghanistan. We know that Nasir al-Din, the

malik of Sistan, had sent an expedition in 695/1295-6 to 'the hot country (garmsir) and the environs of Bust

and Tiginabad' and had cleared the region of 'brigands' (duzdan-u runud).22 Sogedei's forces were possibly

fugitives from the more southerly camping-grounds of the Negiideris, which were in a state of ferment

prior to the advent of Qutlugh Qocha and had become a prey to neighbouring dynasts. Incursions from this

area may have continued into the early years of the fourteenth century. We learn of a second Mongol

assault on Siwistan in 703/1303, coinciding with Taraghai's investment of Delhi (below, p. 223), and a later

one still, which was repulsed in the Than (Thar) region by Alp Khan, the governor of Gujarat, acting in

concert with Tughluq, the muqta' of Deopalpur and future sultan.23

Qaidu died in present-day Kazakhstan in 702/1303 and was succeeded by his eldest son Chapar.

The new sovereign, whose accession did not pass unchallenged, was prepared to be guided by his sponsor

Du'a, who proposed that the Central Asian Mongols recognize Qubilai's successor, the qaghan Temur, and

inaugurate a general peace throughout the Mongol world. The initiative met with a willing response, too,

from his father's enemies. When in 704/1304 Temiir Qa'an's ambassadors arrived at the Ilkhan's court along

with those of Chapar and Du'a to announce the good news, they were welcomed by Kharbanda, who had

recently succeeded Ghazan and now reigned as Oljeitu Sultan.24 As a result of this mission, in which were

represented also the Mongols of Qutlugh Qocha and other subordinate princes,25 the Ilkhan's dominions

were incorporated in the peace established among the rulers of the various Mongol khanates. It was to this

general reconciliation that Khurasan owed the respite from Chaghadayid aggression which it now secured

for almost a decade.

Crisis: the invasions of Qutlugh Qocha and Taraghai

Not so the Delhi Sultanate. In his letter to the qaghan Temur, Chapar had advocated a settlement

in order that the energies of Chinggis Khan's descendants might be released for conflict with their external

enemies, and had mentioned as the specific target of the Central Asian Mongols the

22 Anonymous, Ta'rlkh-i Slstan, 408 (tr. Smirnova, 379).

23 TFS1 Digby Coll. ms., fol. 113b. FS, 288-9 (tr. 463-4). Since Gujarat had just undergone a

second invasion by the Delhi forces (above, pp. 195-6), the Mongol attack occurred no earlier than 1309-

10. 24 See generally Biran, Qaidu, 69-74. I prefer the date given for Qaidu's death by Qashani, 32,

which is supported by Rashid al-Din's statement that the news reached Ghazan in Iraq early in Sha'ban

702/late in March 1303: JT, III, 356 (tr. Arends, 199).

25 Wassaf, 475: the date given here, Jumada I 705/November-December 1305, is too late and,

unless simply an error, must refer to a later embassy than the one which prompted Oljeitu's letter to Philip

IV (Biran, Qaidu, 71 -2).

~territories of Sind and Hindustan.26 For on the frontiers of these regions, no less than of

Khurasan, the forces of Qaidu and Du'a had assumed the direction of military operations, with the result that Mongol pressure on India had considerably increased. Rashid al-Din says of Qutlugh Qocha's forces

that 'they must forever be doing battle with the sultan of Delhi, and the army of Delhi has many times

defeated them'.27 The reign of 'Ala' al-Din Khalji (695-715/1296-1316) witnessed several Mongol

invasions, two of them on a scale far surpassing those of previous decades and threatening the capital itself.

The Mongols appear to have been kept well informed of circumstances within the Sultanate, and it seems

that, just as in their dealings with the Ilkhan, Qaidu and Du'a profited from distractions on 'Ala' al-Din's

other frontiers in order to mount heavy assaults on his empire. Indeed, the sultan's policy of

aggrandizement at the expense of Hindu powers (see chapter 10) afforded them considerable opportunity.

The earliest unequivocal evidence of operations in India by the Central Asian Mongols belongs to 697/1297-8, when Qaidu's noyan Keder28 invaded the Panjab, ravaging the territory as far as the

neighbourhood of Qasur. But Sultan 'Ala' al-Din's brother Ulugh Khan crushed the invaders at a locality

named Jaran Manjur near the banks of the Sutlej on 22 Rabi' II/6 February 1298. The Mongol dead

numbered 20,000, and the prisoners were taken to Delhi to be executed.29 A greater threat was posed by

Qutlugh Qocha's forces. Their first major strike occurred in 699/1299-1300, during the absence of the Delhi

army on the first Gujarat campaign, with which Egyptian sources expressly link it.30 Qutlugh Qocha,

accompanied by

26 Wassaf, 454.

27 JT, II, 173 (tr. Boyle, 142/tr. Verkhovskii, 93).

28 He served under Sarban in Khurasan in 702/1302-3 (Qashani, 18), and is listed by Wassaf, 511,

among those of Qaidu's noyans who crossed the Oxus with Sarban in 706/1306 to submit to the Ilkhan. The

name is almost certainly Mo. keder, 'obstinate', 'quarrelsome': Lessing, Mongolian-English dictionary, 441.

KF, 36, calls his forces 'Qaidu's carrion-eaters' (murdar-khwar).

29 Ibid., 33-6, for the fullest account- a briefer notice in DR, 59-60.TFS, 250, which places the

invasion in 696/1296-7 and does not name the Mongol leader, gives the sultan's commanders as Ulugh

Khan and Zafar Khan. Jaran and Manjur, named in all these sources (though arbitrarily changed to

'Jalindhar' by the editors of TFS), appear from a later reference in TMS, 218, to have lain in the Jalindhar

region: Hodivala, Studies, I, 407; ibid., 246-7, he identifies the two elements as Jagraon and Macchiwara,

respectively S.W. and S.E. of Ludhiana. For Qasur, on the old north bank of the Beah, at 31° 8' N., 74° 28' E., see IG, XV, 149-50.

30 Ibn al-Dawadari, IX, 57; hence Ibn Abi'l-Fada'il, ed. Blochet, 556-7. The link between the

sultan's own plundering campaigns and Qutlugh Qocha's attack is also implicit at Qashani, 189. The

Egyptian sources suggest that it fell in 699. TFS, 254, places it 'at the end of the aforementioned year', i.e.

of the third year of the reign (697-8), which would suggest the late summer of 1298; but the Mongols

surely arrived in the cold season. Bihamadkhani, fol. 386b, erroneously makes the invasion coincide with

the Ranthanbor campaign. Wassaf, 312, gives 694 in error; Rashid al-DIn, JT, ed. Jahn, Indiengeschichte,

Ar. text Taf. 59, Pers. text Taf. 25 (German tr. 50), is vague.

~his brother Temur Buqa,31 advanced directly on Delhi. 'Ala' al-Din met the Mongols at Kili, a

place whose location is now unknown but which apparently lay some fifteen miles north of the capital.32 His right wing, led by Zafar Khan, crushed the Mongol left but on the way back from the pursuit was

ambushed by the enemy rearguard under the noyan Taraghai and annihilated. Yet the Mongol army then

retired.33 Barani's explanation - that their appetite for further conflict had been reduced by the strenuous

resistance of Zafar Khan, whose name was to become a byword among them - is hardly satisfactory.34 The

real reason appears to be that, as we learn from contemporary sources, Qutlugh Qocha had been mortally

wounded: he died during the long return journey to his base.35

Over the next few years Mongol bands numbering 10,000 or 15,000 horse continued to make

plundering raids on the Panjab, but caused no general alarm and retired on each occasion without a pitched

battle.36 But when in 702/1302-3 'Ala' al-Din's forces were again scattered on distant campaigns, Taraghai,

now in command of Qutlugh Qocha's army,37 felt strong enough to threaten Delhi a second time. This invasion appears to have posed an even greater danger than that of Qutlugh Qocha. 'Ala' al-DIn was

reduced to following the defensive tactics he had eschewed during the earlier attack, barricading himself

and his army in the Siri plain. The Mongols' position extended from the Yamuna as far as the plain of

Lohrawat; but although they launched raids into the suburbs of the old city, where they penetrated as far as

the Hawd-i Khass, they were unable to move there in force for fear of exposing their flank. This stalemate

situation lasted for about two months; then Taraghai suddenly withdrew to his own territory.38

31 So called in FS, 260 (tr. 431). This is correct: although he does not appear in JT or in Mu'izz al-

Ansab, he is listed among Du'a's sons in SP, fol. 120a.

32 Hodivala, Studies, I, 271, for a discussion. Of the sources he does not cite, FS, 259 (tr. 430),

says merely that Kili was in the Doab; but TFS1, Bodleian ms., fol. 145a, and RRL ms., 219, makes the

sultan march seven kurohs (1 kuroh - approx. 2 miles) from Delhi to the battlefield. On the strength of the

word-play in DR, 60, the spelling 'Kaili' is advocated in ED, III, 548 n.4; but this is hardly conclusive, as

Khusraw's puns are often just visual.

33 TFS, 260-1. FS, 262-5 (tr. 430-41); ibid., 265-9 (tr. 441-3), 'Isami speaks of a further

confrontation the next day between the Mongols and the sultan's main force. The spelling 'Targhi' usually

found in secondary literature is incorrect: this is Tu. taraqai, 'bald' (Pelliot, Notes on Marco Polo, 69, 568),

and the meaning is confirmed by Khusraw's pun in KF, 37, sar-i asla'. 34 TFS, 261.

35 Qashani, 193, 201. DR, 61, agarchi hall az shamshlr jan burd * wa-lik az sahm-i harba raftana

murd: this detail, which is omitted in ED, III, 548, is reproduced in Bihamadkhani, fol. 387b (with the

verse), TMS, 72-3, and Bada'uni, I, 185. For the meaning of raftana[n], see Hodivala, Studies, I, 268.

36 TFS1 Bodleian ms., fol. 145b/RRL ms., 220.

37 Wassaf, 510 (describing the events of 705/1305-6), lashgar-i Qutlugh Khwajara midanist.

Qashani, 36, calls Taraghai the amir of Qutlugh Qocha's ordo (camp).

38 TFS, 301-2; TFS1, Digby Coll. ms., fols. 113b-l 14a, but Bodleian ms., fols. 145b-146a, omits

the duration of the siege. FS, 285-6, 291-2 (tr. 460-1, 466-7), recounts two invasions by Taraghai, the first lasting forty days and the second one month. He accompanied a later raid, in 1305 (below), when he

certainly did not reach the capital.

~The invasions of Qutlugh Qocha and Taraghai represented crises of the first magnitude. The size

of their armies varies considerably in Barani's accounts, but the lowest figure he gives for Qutlugh Qocha's

force is ten tumens or one lak (100,000); he supplies no statistics for 'Ala' al-Din's army, which may have

been outnumbered, since reports reaching Egypt put it at a mere 30,000.39 In 703/1303 Taraghai may have

brought with him as many as 120,000 men, whereas in one manuscript of Barani's first recension the sultan

is said to have withstood the siege with only 10,000 horsemen and 50,000 foot.40 Although Qutlugh

Qocha's Mongols did not penetrate as far as the outskirts of the city, Delhi felt the impact of the invasion,

since refugees from the surrounding countryside drove up the price of foodstuffs when wary traders were

unwilling to venture near the city.41 At the time of Taraghai's attack, Delhi suffered all the rigours of a blockade. 'Ala' al-Din, himself busy with the reduction of Chitor, had only belatedly realised the magnitude

of the crisis.42 There was no hope of reinforcements: not only were the garrisons at Multan, Deopalpur and

Samana distracted by a Mongol inroad into Siwistan (p. 220 above), but Taraghai had secured all the

Yamuna crossings, so that the divisions of 'Ala' al-Din's forces returning from Tilang were obliged to halt at

Kol and Baran.43

Taraghai's retreat was widely regarded as one of the miracles of the age;44 and certainly the

sources offer no explanation. It may well be that, on the basis of his previous experience of the sultan's

military tactics, he had anticipated a pitched battle and had come unprepared for an investment;45 more

probably, his attention was demanded by events beyond the Oxus, which we shall examine below. His

invasion roused 'Ala' al-DIn to repair various fortresses lying in the path of the Mongol advance: Kaithal was refortified, and an inscription on the Barsi Gate at Hansi enables us to date

39 Ibn al-Dawadari, IX, 57, and Ibn Abi'l-Fada'il, ed. Blochet, 557, for the Delhi army. For the

Mongols, see TFS, 256, although earlier, 254, the figure is twenty tumens; the mss. of TFS1 differ, Digby

Coll. ms. agreeing here with the printed text, while Bodleian ms., fol. 145a, and RRL ms., 219, give the

number of tumens respectively as 'ten or fifteen' and 'ten or twelve'.

40 Ibid., Digby Coll. ms., fol. 114a, for the sultan's forces. For Taraghai's army, see ibid., fol. 113,

with first 'ten or twelve tumens' and then 'twelve'; RRL ms., 220, gives 'one lak and 20,000' (the phrase

employed later in TMS, 73); Bodleian ms., fol. 145b, has 20 or 30,000, suggesting that a phrase has dropped out. TFS, 300, has at one point twelve tumens (but cf. BL ms., with 'two or three tumens') and at

another '30 or 40,000' (BL ms. again differs, with '20 or 30,000'). FS, 285 (tr. 460), has the ludicrously high

figure of 200,000 for the Mongols.

41 TFS, 254-5.

42 TFS1, Digby Coll. ms., fol. 113/Bodleian ms., fol. 145b/RRL ms., 220. Less than a month

elapsed before the Mongol army reached the Yamuna about 10 m. N. of the capital.

43 TFS, 300-1.

44 Ibid., 302. TFS1, Digby Coll. ms., fol. 114a/Bodleian ms., fol. 146a/RRL ms., 221. 45 Biran, Qaidu, 89-90, proposes that Chaghadayid armies were 'relatively unskilled in siege

tactics'. I am not convinced by the argument of Iqtidar Alam Khan, 'Coming of gunpowder to the Islamic

world and North India: spotlight on the role of the Mongols', JAH 30 (1996), 27-45, on the strength of a

single reference in KF, that the Mongols were using gunpowder in siege warfare in India by 1300.

~the restoration here in Rabi' II 703/November 1303.46 The sultan also enacted various fiscal and

administrative measures, designed to increase the armed forces and to avert any repetition of the crisis (see

chapter 12); but however salutary these proved, the people of Delhi had good reason to be thankful also for

the internecine strife which erupted around this time in Central Asia.

The collapse of Qaidu's confederacy and the rise of the Chaghadayids

It was ironic that Chapar, alone of all the Mongol rulers, failed to reap any benefit from the

general peace that he had promoted. His submission to the qaghan placed him on an equal footing with his

erstwhile subordinate, Du'a, who further undermined his position by encouraging the disaffection of various

princes in Chapar's ulus.47 In the war that broke out in 705/1305 Du'a was supported not only by many

Ogodeyid princes, notably Kuresbe and his brothers, but also by the frontier forces of the qaghan in the

east: Chapar was compelled to submit to Du'a and received a much smaller appanage.48 We are concerned

less with these wars as a whole than with their ramifications in Afghanistan and along the upper Oxus,

where the first blow appears to have been struck in the summer of 705/1305 by Taraghai, acting on secret

orders from Du'a. Repulsed by Sarban's forces, he made for India.49 But following an attack by Du'a's son

Esen Buqa, whom his father had sent out to rule Qutlugh Qocha's ulus, Sarban abandoned his bases in Baghlan and in 706/1306, accompanied by Temur, son of Ebugen, and Keder, moved into Khurasan to seek

the protection of the Ilkhan Oljeitu.50 Scattered details in our sources confirm the impression that

Afghanistan was in turmoil. Taraghai's attempt to flee to India in 705/1305, says Wassaf, had been

obstructed by Qutlugh Qocha's wives, who would not let him pass on account of his hostility to Sarban, and

he therefore joined the Neguderis.51 Not long afterwards he was killed when Esen Buqa was obliged to go

to 'Hindustan' - presumably the Indian borderlands - to quell dissension (mukhalafat) within Qutlugh

Qocha's army.52

Conflict continued in Central Asia following the death of Du'a in 1306 and during the brief reign

of his son Konchek, who died in 1308, and peace was restored only in 709/1309 when Esen Buqa was

summoned from Bini-yi Gaw to be khan of an ulus that now comprised not only his father's 46 Wahid Mirza, 'Some Arabic and Persian inscriptions from the East Punjab', EIAPS (1953-4), 8-

9. Mehrdad Shokoohy (ed.), Haryana I (London, 1988), 31-3.

47 Qashani, 33-5. 48 See generally Biran, Qaidu, 73-7.

49 Wassaf, 510. Qashani, 36, having begun to describe this struggle, then abruptly breaks off.

50

Wassaf, 510-11. Qashani, 54, reports the arrival of Sarban and Temur at Oljeitu's court in Rajab

706/Jan. 1307, but according to Wassaf, 512, Sarban remained in Khurasan and died soon afterwards. 51 Ibid., 510, ba-Qara'unas muJhaq shud. 52 Ibid., 517.

~territories but most of Qaidu's also.53 Within a few years, however, this calm was dissipated by

events that had their roots, once more, in the frontier zone of Afghanistan. After the brief rule of Esen

Buqa's younger brother It Qul over the Negiideris,54 we find the region under the control of Da'ud Qocha,

the son or nephew of Qutlugh Qocha. Like Qutlugh Qocha himself, Da'ud Qocha moved between the banks

of the Oxus and 'the furthest parts of Shaburghan', on the one hand, and the hot regions (garmslr) of

Ghazna, Bini-yi Gaw, Bust, Tiginabad and the Indus valley on the other. He proved an energetic ruler,

nourishing designs on Herat and attempting to bring to heel two chiefs, Abachi's sons Temur and *Lakchir.

Since they are described as leading 'the remnants (baqaya) of the Negiideris', it may be that part of the

Neguderi forces had profited from the recent upheavals to escape from the Chaghadayid orbit. At any rate, Temur and *Lakchir sought help from the Ilkhan Oljeitii, whose forces in 712/1312 fell on Da'ud Qocha

and sacked his headquarters at Tiginabad.55

Oljeitu's response to the Neguderi appeal was consistent with the pronounced interest he had

displayed from the outset in his eastern frontier, replacing the local dynasty in Kirman by an Ilkhanid

appointee, reasserting control over Quhistan, and in 706/1307 taking Herat, which for some years had

defied his overlordship. According to reports that reached Egypt the main object of the Ilkhan's ill-starred

campaign of that year to subdue Gilan was to facilitate communications with Khurasan.56 It is tempting to

link this burst of military activity with the embassy which Oljeitii sent to Delhi in 710/1310-11, demanding

the submission of 'Ala' al-Din and the hand of a Khalji princess in marriage. This seems to have been an

isolated contact, however, and it certainly bore no fruit, since the envoys were detained and eighteen members of their suite were crushed beneath the feet of elephants.57

53 Ibid., 513-14, 518-20. QashanI, 147-50. Barthold, Four studies, I, 131-3.

54 QashanI, 150. SP, fol. 120a ('YTQWLY), and Mucizz al-Ansab, fol. 32b, list him among Du'a's

sons. The name is Tu. it, 'dog', + qul, 'slave': Clauson, Etymological dictionary, 34, 615.

55 QashanI, 152-3, 201-2, describing these events twice. Sayfi, 595-8, sub anno 713 and calling

Da'ud Qocha Du'a's son in error: he alone mentions the sack of Tiginabad. Da'ud Qocha is not mentioned in

SP, but appears in Mucizz al-Ansab, fol. 32b, as the son of a brother of Qutlugh Qocha named Qutlugh.

Neguderi chiefs: Qashani, 152; also 201, baqaya-yi Qaraunas-i Nikudari. For the second brother's name

(LKMYR in the printed text), Istanbul ms. Ayasofya 3019, fols. 67a, 89a, reads LKHYR. 56 Kirman: QashanI, 43; Qazwini, Ta'rikh-i Guzida, 536. Quhistan: Qashani, 54. Herat: Sayfi, 461-

97, 503-43; Spuler, Mongolen in Iran, 93-4. Gilan: Ibn al-Dawadari, IX, 149; Ibn Abi'l-Fada'il, ed. Blochet,

641; Boyle, 'Dynastic and political history', 400-1. Cf. Spuler's view of his foreign policy: Mongolen in

Iran, 89-90.

57 Wassaf, 528; this is misrepresented as a friendly embassy in Aziz Ahmad, 'Mongol pressure',

187-8, and his Studies, 16. Akbar's minister Abu'1-Fadl 'AllamI mentions an embassy sent by Oljeitii to

Qutb al-DIn Mubarak Shah Khalji and headed by no less a figure than Rashid al-DIn: A'in-i Akbari, ed. H.

Blochmann, BI (Calcutta, 1872-7, 2 vols.), II, 206; tr. H. S. Jarrett, BI (Calcutta, 1891-4, 3 vols.), Ill, 348.

But it is unlikely that the Ilkhan would have employed such a high-ranking minister for such a mission. On Rashid al-Din's own correspondence, see p. 154 above.

~Da'ud Qocha's flight across the Oxus and his appeal to Esen Buqa unleashed a war between

Chaghadai's ulus and the Ilkhanate which lasted for some years. The forces of the qaghan, whose frontier

had advanced considerably westwards since the collapse of Qaidu's empire, also engaged in hostilities with

Esen Buqa, and the number of embassies between Persia and China suggests that Toluid solidarity had re-

emerged as a factor in the politics of the Mongol world. At one point the Golden Horde too became

embroiled with Chaghadai's ulus.58 The Central Asian Mongols temporarily forfeited control of the

strategic regions of Afghanistan which gave them access to India. We do not even know whether Da'ud

Qocha was reinstated in his old camping grounds.59

When Oljeitu died, his youthful successor Abu Sa'id (716-36/1316-35) was confronted by a revolt

on the part of a renegade Chaghadayid prince, Yasa'ur, who had quarrelled with Esen Buqa and had been

allowed by the Ilkhan to settle south of the Oxus. Yasa'ur threatened Herat and invaded Sistan, where in

717/1317-18 he killed the Ilkhan's adherent, the Negiideri amir Temur. But in 720/1320, before Abu Sa'id's

advancing forces had located him, Yasa'ur perished in an attack by Kebek, Esen Buqa's brother and deputy

in Transoxiana. His ambitions had been sufficiently dangerous to induce his Chaghadayid kinsmen and the

Ilkhans temporarily to collaborate in his removal.60

For a time, it had appeared as if the Ilkhans might exercise authority in eastern Khurasan and

Afghanistan, whether via compliant Negiideri leaders such as Abachi's son Temiir or mediated in the treaty

arrangements with refugee princes from Central Asia like Yasa'ur.61 Oljeitu's death, however, followed by the elimination first of Temiir and then of Yasa'ur, facilitated the revival of Chaghadayid power here. This

fresh advance may date from an invasion of eastern Persia in 722/1322 by Kebek, who had now followed

Esen Buqa as khan of Chaghadai's ulus (c. 718-726/1318-1326).62 His

58 Spuler, Mongolen in Iran, 97-8. Barthold, Four studies, I, 133. T. T. Allsen, 'The Yuan dynasty

and the Uighurs of Turfan in the 13th century', in Morris Rossabi (ed.), China among equals: the Middle

Kingdom and its neighbours, 10th-14th centuries (Berkeley and Los Angeles, 1983), 259-60. For the

participation of the Golden Horde, see MA, ed. Lech, 79 (German tr. 144-5).

59 The anonymous continuation (dhayl) of JT, Istanbul ms. Nuruosmaniye Kutuphanesi 2799 (old

numbering: 3271), fol. 25b, claims that he was, but is not supported by Qashani or by Sayfi. 60 Russell G. Kempiners, Jr, 'Vassaf's Tajziyat al-Amsar wa Tazjiyat al-A'sar as a source for the

history of the Chaghadayid khanate', JAH 22 (1988), 178, 185-6. Kato Kazuhide, 'Kebek and Yasawr - the

establishment of the Chaghatai Khanate', Memoirs of the Research Department of the Toyo Bunko 49

(1991), 97-118. Spuler, Mongolen in Iran, 98-101. For Temiir, see Sayfi, 677.

61 Kempiners, 'Tajziyat al-Amsar’, 184-5.

62 al-'Ayni (d. 855/1451), Ta'rikh al-Badr fi Awsaf Ahlil-'Asr, BL ms. Ar. 985 (Add. 22360). fol.

13a; also in V. G. Frhr. von Tiesenhausen, Sbornik materialov otnosiashchikhsia k istorii Zolotoi Ordv, I

(St Petersburg, 1884), Ar. text 494, tr. 524-5. Cf. the anonymous dhayl of JT, fol. 57a.

~successor Tarmashirin (726-35/1326-34), yet another of Du'a's sons, was attacked by the Ilkhan's

forces and defeated in the region of Kabul and Zabul in c. 726/1326; Ghazna was sacked.63 But the reverse

did not, it seems, bring about a change of masters. Ibn Battuta, passing through Chaghadai's ulus on his

way to India in 733/1332-3, refers to Ghazna as part of Tarmashirin's dominions, although it was largely in

ruins. It was subject to the khan's chief amir Boroldai, who was based at Parwan in the Hindu Kush but had

his officers (nuwwab) in Ghazna.64 The indications are that the Ghazna region remained within the

Chaghadayid sphere of influence until the rise of Termir-i Lang.

The later incursions

Inroads into the subcontinent by the armies of Du'a and Qaidu did not cease during the upheavals that followed the latter's death. Early in 705/in the autumn of 1305 Du'a's forces under 'A1i Beg65 and

*Tartaq entered India. Undeterred by the desertion of Taraghai, who turned back after crossing the

Jhelam,66 they pushed deep into the Panjab, ravaging the Siwalik foothills, and then overran Bada'un and

Awadh. The akhurbeg Malik Nanak, who held the iqta's of Sunnam and Samana and who was accompanied

by a number of other amirs, including Tughluq, routed the invaders on 12 Jumada 11/30 December in the

neighbourhood of Amroha. 'A1I Beg and *Tartaq were taken to Delhi, but their lives were spared and they

were kept for a time in honourable captivity.67

The details of the last Mongol attacks of 'Ala' al-Din's reign are somewhat blurred. It seems that

Amir Khusraw, writing only a few years afterwards, describes one invasion by an army comprising three main divisions, and that subsequent authors, beginning with Barani, misinterpreted him and assumed that

there were a number of separate incursions, each falling in a different year. The Mongols were apparently

under the

63 Qazwini, Ta'rikh-i Guzida, 617.

64 IB, III, 42, 82-3, 87 (tr. Gibb, 561, 585-6, 589). Aubin, 'Khanat de Cagatai', 17-18. For Ghazna's

ruined state, see IB, HI, 88 (tr. 590).

65 'Ali Beg, called a descendant of Chinggis Khan in TFS, 320, belonged in fact to the Qonqurat

tribe and was married to a Chinggisid princess: hence the style kuregen, 'son-in-law', given him by Wassaf,

526. His wife was a great-grand-daughter of Ogodei: SP, fol. 127a, adding aknun dar Dilli ast; hence Mu'izz al-Ansab, fol. 42b, ba-Dilli raft. Wassaf confirms that his troops were Du'a's (tua'i).

66 KF, 37-8, sahm-i baylak-zanan-i ghuzza dar dil gudhardnld wa-ham az 'aqab khala kard ('he let

the ... arrows of the ghazis pierce his resolve, and turned about'). This figurative phrase has been

misinterpreted to mean that he was killed, e.g. by ED, III, 72, and Lal, History of the Khaljis, 144.

67 Wassaf, 527 (with the year 708 in error). KF, 38-9, supplies the date. Only TFS, 320, specifies

the locality. See also FS, 303-5 (tr. 479-82); Wassaf, 526-7. One ms. of KF reads NAYB in error for NANK,

and later authors duly name the sultan's general on this occasion as Malik Na'ib (Kafur): thus TMS, 73, and

Bada'uni, I, 185, who wrongly equates Malik 'Manak' (above, p. 175) with Malik Na'ib.

~overall leadership of Kopek, who commanded Du'a's forces south of the Oxus,68 and his two

colleagues are named as Iqbal and Taibu. Entering the Sultanate in the vicinity of Multan and plundering

along the banks of the Ravi, the Mongols moved on Kuhram and Samana, but then turned south towards

Nagawr.69 Malik Kafur 'Hazardinari, Tughluq, and other amirs were sent against the invaders, who were

surprised near a river which Khusraw calls the "Ali-Wahan' but which figures in Barani's account as the

Ghaggar. The Mongol vanguard was completely routed, and Kopek taken prisoner. Kafur then crushed the

forces which were following at some distance, and Iqbal and Taibu fled back across the Indus.70

Barani then furnishes details of two other incursions. First, three or four tumen-commanders

invaded the Siwalik region, but the Delhi army occupied the river-crossings and cut off their retreat.

Having extended their lines of communication deep into a waste country, the Mongols were easily

overcome. On the sultan's orders, the survivors were massacred in the fortress of Nara'ina. Lastly Iqbal, whom Barani calls Iqbalmanda, invaded India, but was defeated and killed in the vicinity of the 'A1I-

Wahan.71 The term 'Siwalik', which in its broadest sense embraces the territory from the foothills down to

Nagawr; the reference to Nara'ina, not far east of Nagawr; and the recurrence of 'Ali-Wahan - all these

details suggest that the three episodes noticed in the Ta'nkh-i Firuz-Shahi were in reality part of the same

invasion as recounted by Amir Khusraw.72

Although they penetrated more deeply into the sultan's territories, and in 1305, at least, beyond the

Ganges, these later attacks posed less of a menace than those of Qutlugh Qocha and Taraghai. In each case

the Mongol army appears to have been smaller. 'Ali Beg and *Tartaq brought 50,000 horse, according to

Khusraw, though other sources supply lower numbers.73 The figure of 100,000 given by 'Isami for the army

of Kopek and his colleagues 68 JT, II, 570 (tr. Boyle, 313/tr. Verkhovskii, 202). FS, 318, calls him sar-ahang-i an kishwar ('the

vanguard of that country'). He was with Sarban in Khurasan in 1302-3: Qashani, 18.

69 KF,A2. DR, 61.

70 They are listed with Kopek in the brief account ibid., 61-2. Fuller narrative in KF, 43-4. The

correct form of Taibu's name, which appears as TABW or TYHW in DR and as TAYBW in KF, is discussed by

Hodivala, Studies, I, 248, 372. It could represent either tabu, 'five', or tayibu, 'quiet', 'calm': Lessing, 761,

767. The river 'Ali-Wahan is mentioned also in FS, 319 (tr. 496); see TFS, 321, for the Ghaggar. 71 Ibid., 321-2. The word before cAli-Wahan, which in the printed text reads TNBDH, proves, on

comparison with the mss., to be a corruption of DHNDH. This was assumed to be the Dhandh: see Hodivala,

Studies, I, 397. But the word occurs in TFS1 in a quite different context (Bodleian ms., fol. 137b/Digby

Coll. ms., fol. 103b), as a synonym for dih, 'village'. Barani also inserts amir before 'Ali-Wahan, and AHG,

II, 816-17, explains that Amir 'A1i commanded at Dhandh as a subordinate of Tughluq. We cannot rely on

this, given AHG's frequent errors regarding earlier invasions.

72 The verdict of Lal, History of the Khaljis, 147-9.

73 KF, 38. Wassaf, 526, has three tumens, though this is difficult to reconcile with the figure of

60,000 for the heads of slain Mongols. TFS, 320, has 30 or 40,000: of the mss. of TFS1. Digby Coll. ms., fol. 121b, gives the same figure, while Bodleian ms., fol. 146b, and RRL ms., 221, have simply 'several

thousand'.

~is clearly exaggerated, in view of the fact that they turned south from Samana because they

lacked the strength to proceed further.74 Barani's reference to 'three or four tumen-commanders',75 although

it occurs in a passage which is probably confused, gives a more realistic idea of the size of these invading

armies. By this time the Mongol heartlands on the upper Oxus and in Transoxiana were torn by civil war.

To some extent organized expeditions may have been superseded by the inroads of fugitives seeking to

settle on a more permanent basis, as was happening both in Khurasan and on the Chinese frontier.

As a result of 'Ala' al-Din's victories, says Khusraw, the Mongols withdrew into 'the mountains of Ghazna' and were unable to pass through Sind.76 During these years the Delhi forces may have moved over

to the offensive. According to Barani, Ghazi Malik Tughluq, who had at some point received the additional

iqta' of Lahore, not only kept the Mongols at bay but took the offensive against them, heading an

expedition every winter into their territory. Later he credits Tughluq with twenty victories over them.77 No

details are supplied, and his assertions might be questioned were it not for other testimony regarding

Tughluq's exploits. Amir Khusraw, in his Tughluq-Nama, written to commemorate Tughluq's enthronement

in 720/1320, alludes to eighteen victories, mostly over the Mongols; while Ibn Battuta saw an inscription in

the mosque at Multan in which Tughluq himself laid claim to twenty-nine victories over the Mongols

alone.78 Whether these campaigns were responsible for the devastation of an extensive tract between

Ghazna and India, which al-'Umari's informants attributed to the strife between the 'king of India' and the

'king of Turkestan and Ma wara' al-Nahr',79 is uncertain.

That the Sultanate now enjoyed immunity from major Mongol attacks for some years was due in

large measure to conditions in Afghanistan, which are momentarily but vividly illuminated for us by a

remarkable document preserved in Amir Khusraw's Rasa'il al-I'jaz.80 It purports to be a memorial ('ard-

ddsht) from the chamberlain (hdjib) Badr to Sultan 'Ala' al-Din's son Khidr Khan, narrating a winter

campaign against the Mongols of Ghazna. The Delhi forces, led by an unnamed grandee who is designated

simply as the khdn-i a'zam, had allegedly occupied the city of Ghazna,

74 KF, 42. FS, 318 (tr. 495), for the alleged 100,000.

75 TFS, 321.

76 KF, 113.

77 TFS, 322-3; and see 416 for the twenty victories. Later, 490, Barani ascribes these twenty

victories to Tughluq and his brother (Rajab) - doubtless in an effort to curry favour with Rajab's son Firuz

Shah.

78 Tughluq-Nama, 138. IB, III, 202 (tr. Gibb, 649).

79

MA, ed. Spies, 8 (German tr. 30)/ed. Fariq, 16 (tr. Siddiqi and Ahmad, 32).

80 RI, IV, 144-56; for a brief abstract, see ED, III, 566-7. The document has been studied by

Askari, 'Material', 18-20, and by M. Y. Z. Siddiqi, 'Arzdasht of Badr Hajib', MIM 2 (1972), 291-7.

~where the khutba was read in 'Ala' al-Din's name.81 Badr's memorial has justifiably aroused

considerable suspicion among scholars.82 Such a major triumph as the capture of Ghazna - uncorroborated

in any other source - is improbable. The document nevertheless contains enough circumstantial detail to

suggest that it is based on a genuine intelligence report from the north-west frontier to the sultan's

government.83 Badr refers to the fratricidal war that was raging between Du'a and Qaidu's people,84 and

describes how it had spread to the Mongol army based at Hashtnaghar and Peshawar, with the result that

chaos reigned between Ghazna and the Indus.85 He goes on to say that Esen Buqa had moved north in

response to a message from Konchek: before his departure, he had presented Badr with several gifts for his

master Khidr Khan, by way of a conciliatory gesture. The memorial may safely be dated, therefore, to the years 706-7/1306-8, when Konchek was head of Chaghadai's ulus.86 For all the problems attached to it, the

document does at least furnish first-hand evidence that the principality built up by Qutlugh Qocha south of

the Oxus had begun to disintegrate.

Defeat meant for many Mongol warriors an unpleasant form of execution. The crushing of

Mongol captives beneath the feet of elephants has made 'Ala' al-Din notorious, although these tactics are

first encountered in the reign of Mu'izz al-Din Kayqubad.87 Following the incursion by 'Ali Beg and

*Tartaq a durbar was held in which captured Mongols were executed in this fashion as a spectacle for the

citizens of Delhi; and nemesis took the same form subsequently both for Kopek and for Iqbal's officers.88

An unsavoury practice which does date from around this time was the construction of towers at Delhi with

the heads of the slain. According to Barani, such a tower could still be seen in front of the Bada'un Gate in his own day;

81 RI, IV, 148, 150-1.

82 See, e.g., Day, 'North-west frontier', 106-7 n.20, and in his Some aspects, 55 n.22. Askari, 'Risail

[sic]-ul-Ijaz', 122, appears to believe the report is Khusraw's own invention.

83 See Khusraw's own comment on the document: RI, IV, 18.

84 Ibid., IV, 151-2. The printed text reads: DWAYR Ia'in TBR QYD W MY AN gardunan-i kafir tigh

uftada ast. But for the first word, IOL Persian ms. 570 (Ethe, no. 1219), fol. 223b, and BL ms. Add. 16842,

fol. 404b, read DW, and the subject can only be Du'a 'the Accursed' (Ia'in). The next few words are problematic: a line is possibly omitted not only in the mss. but -since Khusraw was himself transcribing a

document - in the original, QYD W MY AN can only be Qaydu'iyyan, 'Qaidu's people'.

85 Ibid., IV, 153-4 (HYBT NPYR to be corrected to HSTNTR from IOL ms., fol. 224a, and BL ms.

Add. 26841, fol. 382b); for Hashtnaghar, 16 m. N.W. of Peshawar, see IB, tr. Gibb, 591 n.212.

86 RI, IV, 154, 155. The Mongol leader's name appears both here and ibid., 147, as 'YS BrA, a form

which prevented Askari ('Material', 18 n.50) from identifying him; but the best mss. have clearly 'YSN BrA.

His name is Tu. esen, 'healthy', + buqa, 'bull': Clauson, Etymological dictionary, 248, 312. Konchek's name

appears as QPCK. The margin of 1307-14 allowed by Siddiqi, 'Arzdasht', 292, is unnecessarily wide.

87 QS, 96-8: on this occasion it was only one among several kinds of grisly death on offer.

88 TFS, 321, 322'. FS, 322, on the other hand, says that Kopek was initially spared and was later

beheaded (verses omitted in Husain's tr., 500).

~although the sources differ as to whether it was built from the skulls of Kopek's soldiers or those

of 'Ali Beg.89 No doubt successive invasions permitted the tower to be completed in two stages. If we are to

believe Khusraw, who dwells on the marauders' fate with particular pleasure, similar towers arose in other

cities of the Sultanate and Mongol prisoners' remains were also incorporated in the new fortifications at

Delhi.90 Whether such exercises were as effective in deterring the Mongols from future inroads as they were in entertaining the populace of the capital, we cannot tell.

Barani asserts that the respite from Mongol attacks lasted until the end of the reign of Qutb al-Din

Mubarak Shah (720/1320);91 and Khusraw claims that Qutb al-Din contemplated the conquest of Ghazna

but was dissuaded by his amirs.92 But Barani's statement that the Mongols had been so cowed by Tughluq

as not to dare to invade India during his reign93 is contradicted by his own evidence, for he tells us of one

raid which occurred shortly after the Deccan campaign of 721/1321-2. A fuller account of this attack is

provided by 'Isami, who says that the sultan sent reinforcements to his nephew and lieutenant at Samana,

Baha' al-Din Garshasp. Garshasp attacked the Mongol rearguard, which had remained behind at a base

camp in the foothills, and routed them, slaying their commander, *Shir (Shira). Thereupon he ambushed

and destroyed the rest of the invading army, under three commanders named Hindu, Zakariyya and Orus,

close to the left bank of the Beah as they returned from plundering the Doab. Among the prisoners was Zakariyya, who was taken to Delhi in triumph.94

Muhammad b. Tughluq and the Mongols

Soon after his accession (724/1324), Muhammad b. Tughluq headed an expedition to the north-

west. While the sultan halted at Lahore, his troops took Kalanawr and Peshawar and had the khutba read

there in his' name. Within a few weeks, Muhammad's generals were obliged to retreat owing to lack of

grain and fodder, and rejoined the sultan, who remained at Lahore for two or three months in order to

pacify the region before returning to Delhi. It may have been at this time that the sultan set in motion the

repair of the fortress of Kalanawr, which appears in al-'Umari's incomplete list of his territories.95 Another

consequence of this campaign seems to have been 89 Kopek's troops: TFS, 321; KF, 45-6. 'A1i Beg's: FS, 305 (tr. 481); Wassaf, 527. Towers had

been built earlier from the heads of slaughtered Hindus: Taj, fol. 137b. 0 KF, 28. RI, I, 17, dar aqasi-yi

mamalik niz burjhd-yi digar ham bar In nahj sarasar ba-ras-I falak rasanidand.

91 TFS, 322, 323; cf. also 387.

92NS, 54-5.

93 TFS, 441.

94 FS, 405-9 (tr. 611-18). The brief account in TFS, 450, says that two Mongol leaders were

captured.

95 FS, 423-4 (tr. 649-50). TMS, 101, for Kalanawr, though dating its restoration in the wake of

Tarmashirin's invasion. MA, ed. Spies, 6 (German tr. 26)/ed. Fariq, 14 (tr. Siddiqi and Ahmad, 30).

~the incorporation of the Peshawar region into Muhammad's dominions, since Ibn Battuta

describes Hashtnaghar as 'the last inhabited place on the confines of the land of the Turks [i.e. the

Mongols]' and elsewhere indicates that it was a frontier post where the sultan's customs officals levied duty

on imported horses.96

Yet a third effect of Muhammad's expedition was to bring down upon the Sultanate an invasion by the khan Tarmashirin, who resided in the western half of Chaghadai's ulus, possibly at Tirmid on the

middle Oxus. The authenticity of this invasion was long doubted, on the grounds that it is not mentioned in

the standard text of Barani's Ta'rikh-i Firuz-Shdhi, and it was suggested that Tarmashirin paid a friendly

visit to India to seek Muhammad's assistance. The unearthing of an earlier recension, however, in which the

invasion is in fact described, has undermined this hypothesis.97 Tarmashirin clearly profited from the poor

state of frontier defence in the wake of Kushlii Khan's revolt at Multan (below, p. 257).98 At a date which

Barani places within the two or three years after Delhi's citizens had been transferred to Dawlatabad, and

which Sirhindi gives as 729/1328-9,99 the Chaghadayid forces overran a considerable area, capturing

several fortresses and taking prisoners throughout the regions of Lahore, Samana and Indri. They then

advanced into the Doab. The sultan mustered a large force, which he stationed north of Delhi, setting up his headquarters at Indrapat, close to the Yamuna,100 so that unlike 'Ala' al-Din in 703/1303 he controlled at

least one of the crossings. A division of his forces under Yiisuf-i Bughra, who had been sent to relieve

Mirat, routed part of Tarmashirin's army and captured his nephew. The Mongols shortly withdrew,

followed by Muhammad and his army: Isami says that the sultan halted at Thanesar and sent troops in

pursuit; Sirhindi, that he advanced as far as Kalanawr.101 This was the last major Chaghadayid invasion

prior to Temur's conquest of Delhi at the end of the century.

96 IB, II, 373, and III, 90 (tr. Gibb, 478, 'Shashnaqar', 591, 'Shashnagar').

97 Jackson, 'The Mongols and the Delhi Sultanate', 119-26, surveys the evidence. For the older

view, see especially A. M. Husain, The rise and fall of Muhammad bin Tughluq (London, 1938), 100-8,

and Tughluq dynasty, 119-43. 98 TFS,479.

99 TFS1 Digby Coll. ms, fol. 160b/Bodleian ms., fol. 192a. TMS, 101.

100 Best text in TFS1, RRL ms., 287; cf. Bodleian ms., fol. 192a/Digby Coll. ms., fols. 160b-161a.

Bihamadkhani, fol. 400a, specifies Indrapat. This attempt to control the crossing seems more plausible than

the statement in FS, 463 (tr. 698), that Muhammad's army stretched from Siri to the Bagh-i Jud.

101 Ibid., 463-5 (tr. 699-701). TMS, 101. A less reliable account of the episode is furnished by a

Timurid chronicler, who alleges that Tarmashirin was bought off by the sultan and that he ravaged Gujarat as he withdrew: Yazdi, ZN, ed. A. Urunbaev (Tashkent, 1972), fols. 8Ob-81a. For his failure at Mirat, see

Ghiyath al-Din 'Ali Yazdi, Ruz-Nama-yi Ghazawat-i Hindustan, tr. A. A. Semenov, Dnevnik pokhoda

Timura v Indiiu (Moscow, 1958), 129, 131; Shami, ZN, I, 194; Yazdi, ZN, ed. M. M. Ilahdad, BI (Calcutta,

1885-8), II, 129, 132/ed. Urunbaev, fols. 328a-329a.

~Muhammad endeavoured to form a coalition against the Chaghadayids with the Ilkhan Abu

Sa'id. A local historian writing in southern Persia a few years later has transmitted an account of a friendly

embassy from the sultan to the Ilkhan in 728/1327-8, and the letter it conveyed, in which Muhammad

sought military collaboration against Tarmashirin, has also survived.102 Nothing came of these negotiations,

as far as we are aware, apart from an equally cordial reply from the Ilkhan, but Muhammad allegedly

continued to send annual embassies down to Abu Sa'id's death.103 Subsequently, according to Ibn Battuta,

Muhammad and Tarmashirin too were on friendly terms and exchanged letters and gifts. Professor Siddiqui ascribes this to the khan's conversion to Islam, which seems to have occurred after 729/1328-9.104 But the

reconciliation may be connected with his conflict in the eastern half of the khanate with his brother Dore

Temur, who was overthrown late in 1331.105 Within a few years, Tarmashirin himself was overthrown and

killed, ushering in a period of instability within the Chaghadayid khanate. The new khan, Dore Temur's son

Buzun, though described by Ibn Battuta as a Muslim, apparently preferred the Mongol customary law, the

Yasa, to the Shari'a, and was in any case unable to establish his authority before he in turn was displaced in

1335 by a pagan cousin, Changshi, who was hostile to Islam.106 These upheavals, al-'Umari was informed,

provided the Sultanate with a respite from Mongol attacks.107

His co-religionists' plight furnished Muhammad with an alternative means of extending his

influence beyond the Indus. The manshur drafted on his behalf in 734/1333-4 and sent to Transoxiana to invite sayyids, shaykhs, 'ulama', bureaucrats and soldiers to come to India and enter the sultan's service is

preserved in the early fifteenth-century insha collection, Fara'id-i

102

Shabankara'i, 287-8, naming Ikhtisan (above, p. 153) among the party; brief mention in Fasih-i

Khwafi, Mujmal-i Fas'ihi (845/1441-2), ed. Mahmud Farrukh (Tehran, 1339 Sh./ 1960, 3 vols.), Ill, 39.

Bayad-i Taj al-Din Ahmad Wazir, ed. Iraj Afshar and Murtada Taymuri (Isfahan, 1353 Sh./1974), 404-8,

for the letter; also I. H. Siddiqui, 'Sultan Muhammad bin Tughluq's foreign policy, a reappraisal', IC 62

(1988), part 4, 10-12. The details of another embassy by Ikhtisan to Abu Sa'id (dating his return at the time

of Muhammad's death!) in Bihamadkhani, fol. 405b, are chronologically dubious.

103 Shabankara'i, 88, 288 (naming the ambassador as cAdud al-Din of Yazd: see above, p. 184).

For the Ilkhan's reply, see Bayad, 408-9.

104 IB, III, 43 (tr. Gibb, 562). Siddiqui, 'Sultan Muhammad bin Tughluq's foreign policy', 14. MA,

ed. Lech, Ar. text 38, simply dates his conversion as 'since 725' (tr. 117 has 'seit 750' in error).

105 Yuan Shih, cap. 35, cited in MA, ed. Lech, 241 n.167.

106 Barthold, Four studies, I, 135-6. al-Safadi, Waft, X, 383, dates Tarmashirin's death in 735. For

Buzun, see Yazdi, ZN, ed. Urunbaev, fol. 81a. MA, ed. Lech, Ar. text 22 (German tr. 105), describes the

Chaghadayid khanate as in a state of upheaval (mutakhabbatan) until Changshi's accession, for the date of

which see Aubin, 'Khanat de Cagatai', 24-5 n.34. 107 MA, ed. Lech, Ar. text 40 (tr. 118 renders wa-ikhtalafat kalimat ahl hadhihi'l-bilad 'ala

mulukiha as 'Die Inder waren mit ihren Fiirsten uneins'; but the subject is the people of Chaghadai's ulus).

~Ghiyathi.108 That it was effective is clear from the number of notables from Transoxiana who

arrived in the Sultanate at approximately the same time as Ibn Battuta (above, p. 185). It is in this context

too that we must place the fresh influx of Mongols into the Sultanate. Soon after Buzun's seizure of power,

Muhammad had welcomed Tarmashirin's son *Pashaitai, his daughter and her husband Nawruz Kuregen;

and within a short time the number of Mongols from Tarmashirin's dominions in the Sultanate is set by Ibn

Battuta at 40,000.109 In the 1340s refugees were arriving in great numbers. Every winter, according to

Barani, Mongol commanders of tumens and hundreds, their wives (khatunan) and sons (ughliyan) arrived in India and received presents of money, jewels and horses.110 Elsewhere the same author says that

Muhammad caused any amirs from Khurasan and Mughalistan who entered his service to take an oath of

allegiance with the caliphal diploma prior to the conferment of gifts.111 Since the 'Abbasid embassy did not

reach Delhi until 744/1343-4 at the earliest (see below, p. 272), this indicates that immigrants from the

Mongol world were still arriving several years after Tarmashirin's death. But the greater significance of

Barani's information is that these immigrants were Muslims; otherwise oaths involving the caliphal

diploma would have been meaningless. Their flight from Transoxiana may have been connected with the

overthrow, in c. 743/1342, of the ephemeral Muslim khan Khalil, allegedly a son of Yasa'ur (above, p.

226), an obscure episode for which the somewhat dubious account furnished by Ibn Battuta is regrettably

our sole evidence.112

Muhammad profited from the disturbances within Chaghadai's ulus and used the enormous patronage at his disposal in order to cement harmonious relations with Mongol chiefs and other rulers in

Khurasan. In his first recension, Barani goes so far as to claim that 'the whole of Mongol territory

(Mughalistan) on this side of Transoxiana became Sultan Muhammad's obedient client (banda-yi

parwarda)'.113 If Ibn Battuta is to be believed, even Mu'izz al-Din Husayn, the Kartid malik of Herat, at

some point became

108 FG, SK ms. Fatih 4012, fols. 456a-457b; at fol. 457a he refers to their tribulations. Aubin,

'Khanat de Cagatai', 22.

109 IB, III, 43, 46 (tr. Gibb, 562, 564). For Nawruz, see also TFS, 533; SFS, 4 (tr. Basu, JBORS 22

[1936], 96). The name of Tarmashirin's son appears as 'Bashai' in the mss. of IB, but the form in Mu'izz al-Ansab, fol. 32a, suggests Bashaitai, 'man of the Pashai', a people and region of the Hindu Kush and situated

roughly N. of Kabul: see Pelliot, Notes on Marco Polo, 799-800.

110

TFS, 499; and cf. also TFS1, Bodleian ms., fol. 199/Digby Coll. ms., fols. 165b-166a.

111 TFS,494-5.

112 IB, III, 48-51 (tr. Gibb, 565-7). On this, see W. Barthold, Zwolf Vorlesungen iiber die Turken

Mittelasiens, tr. Th. Menzel (Hildesheim, 1935; repr. 1962), 206-7; Jurgen Paul, 'Scheiche und Herrscher

im Khanat Cagatay', Der Islam 67 (1990), 284-91.

113 TFS1, Digby Coll. ms., fol. 166a; the text in Bodleian ms., fol. 199b, is corrupt here and reads

WBRDH for parwarda. TFS, 505, speaks more vaguely of the homage of the rulers (dabitan) of Mughalistan.

~Muhammad's client: the malik was evidently concerned that his adoption of the style of sultan in

750/1349 should not prejudice his relations with Delhi.114 Possibly Husayn's usefulness lay partly in his

recently acquired control over the Mongols of the puppet Ilkhan Togha Temur, who from their bases in the

Herat region were in the habit of raiding India.115 By the end of his life, Muhammad had entered into

amicable relations with the noyan Qazaghan, who since 747/1346-7 had been the real power in the western

half of Chaghadai's ulus. Qazaghan, a Muslim of Qara'unas stock (as the Tughluqids themselves may have

been: above, p. 178), furnished him with 4-5000 Mongol troops for his final campaign against rebels in

Sind.116

Plunder or conquest?

Regarding the war-aims of the Mongols, the sources are ambivalent. It is significant that Amir

Khusraw neglects to mention the invasions of Qutlugh Qocha and Taraghai in his Khaza'in al-Futuh,

presumably because neither episode redounded to 'Ala' al-Din's credit. On both occasions the Mongols

appeared in large numbers; on both occasions they advanced by forced marches and caught the sultan

perilously off guard. Qutlugh Qocha's forces are expressly said to have abandoned their usual practice of

plundering the territory on their route.117 One or two contemporary references indicate that Qutlugh Qocha

and Taraghai actually aimed to conquer the Sultanate,118 and the idea has been taken up in turn by

historians writing in this century.119 A policy of long-term conquest might well have been explained by Aziz Ahmad's proposal that the Chaghadayids sought an outlet in India because they were restive under

Qaidu's tutelage and were caught, moreover, between the armies of the qaghan in the east and of the

Ilkhans.120 It is true that the late thirteenth-century Chaghadayid khanate presents an appearance of being

largely hemmed in by other Mongol states, and that Rashid al-DIn makes the khan Baraq, for instance,

complain at the quriltai of 667/1269 that his ulus, in contrast with those of his kinsmen, had no

114 IB, III, 74 (tr. Gibb, 580). FG, I, 146-9, 182-5, and SK ms. Fatih 4012, fols. 196b-197b. Aubin,

'Khanat de Cagatai', 32-3. Siddiqui, 'Sultan Muhammad bin Tughluq's foreign policy', 19.

115 IB, III, 70, 71 (tr. Gibb, 578); on him, see P. Jackson, 'Togha Temiir', Enc.Isl2.

116 TFS, 524 For Qazaghan, see Beatrice Forbes Manz, The rise and rule of Tamerlane (Cambridge, 1989), 33-4, 43-4. Bihamadkhani, fol. 328b, speaks of their friendship.

117 TFS, 254, giragir; cf. TFS1, Digby Coll. ms., fol. 94a, kuch ba-kuch-i mutawatir. But the later

SFS, 187 (tr. Page, 34), refers to Qutlugh Qocha's forces entering the Topra region, near the Sirmur hills.

118 Qutlugh Qocha: Wassaf, 312; JT, ed. Jahn, Indiengeschichte, Ar. text Taf. 59, Pers. text Taf. 25

(German tr. 50); Qashani, 189. Taraghai: FS, 292 (tr. 467). See also AHG, II, 796; Bada'uni, I, 184.

119 Haig, in Cambridge history of India, III, 102. Dharam Pal, 'cAla'-ud-Din Khilji's Mongol

policy', IC 21 (1947), 258. Lal History of the Khaljis, 134. HN, 338.

120 Aziz Ahmad, 'Mongol pressure', 186, and in his Studies, 15. Cf. also Pal, 'cAla'-ud-Din Khilji's

Mongol policy', 257.

~room for manoeuvre.121

Yet the idea that such constriction explains the expansionism of later

decades runs counter to the evidence in three respects. Firstly, Du'a and Qaidu appear to have acted in close

cooperation right down to the latter's death. In the second place, the Ilkhan's eastern provinces offered as

fruitful an opportunity for expansion as did the Indian borderlands; the Ilkhans were, if anything, on the

defensive and Ghazan at least evinced a far greater interest in the frontier with the Mamluks. Nor, thirdly, is

there reason at this stage to posit pressure from the Far East. Qubilai had jettisoned his policy of expansion

in Central Asia during the 1280s - that is, before Qaidu and his allies began to assert their control over the borderlands of India - and Uighuristan was abandoned to the Chagha-dayids by c. 1300.122

The Mongol invasions of India in 699/1300 and 703/1303, therefore, were probably no more than

plundering expeditions on the scale necessary for such a formidable objective as Delhi; though even then

the purpose may have been to 'soften up' the region as a preparation for campaigns of conquest in future

years. The evidence, regrettably, does not permit us to go further. We need not be influenced, incidentally,

by the presence, on a number of occasions, of women and children in the Mongol armies. Thus women are

said to have ridden with Qutlugh Qocha during his march on Delhi; *Sogedei's troops were taken prisoner

with their wives and offspring; and of the 18,000 prisoners who fell into the hands of Alp Khan at Thari,

some 3000 were women. Not long afterwards, more women and children were spared and sent to Delhi to

be sold in the slave markets when their menfolk were massacred at Nara'ina.123 None of this need surprise

us. Thirteenth-century European and Chinese sources show Mongol women riding to war alongside their menfolk.124 In any case, if the Mongol attacks were essentially seasonal migrations between summer

pastures in the uplands of Ghur and Ghazna and winter quarters in the Panjab and beyond, then we should

expect the entire 'horde' to be on the move rather than just the male warriors.

It is already clear in the thirteenth century that the Mongol campaigns in India were designed to

amass great numbers of slaves.125 Sali Noyan's

121 JT, III, 110-11 (tr. Arends, 72).

122 Dardess, 'Mongol empire to Yuan dynasty', 142-3. Allsen, 'Yuan dynasty and the Uighurs of

Turfan', 255, 258-9. 123 Qutlugh Qocha: FS, 256 (tr. 427). Sogedei: TFS, 254. Thari: FS, 289 (tr. 464). Nara'ina: TFS,

321-2. Note also women and children with 'Abd-Allah's invading army in 691/1292: ibid., 219.

124 Paul Ratchnevsky, 'La condition de la femme mongole au 12e/13e siecle', in Walther Heissig et

al. (eds.), Tractata Altaica Denis Sinor sexagenario ... dedicata (Wiesbaden, 1976), 511-Thomas

Spalatensis, 'Historia pontificum Salonitarum', in A. F. Gombos (ed.), Catalogus fontium historiae

Hungaricae (Budapest, 1937-43, 4 vols. with continuous pagination). 2236-7.

125 Lahore (639/1241): TN, II, 165 (tr. 1135). Numerous captives, both Muslim and Hindu, left

behind in 643/1245-6: ibid., II, 55-6 (tr. 813).

~Campaigns in Kashmir and India yielded Hulegu a great booty in Indian slaves, according to

Rashid al-Din, who says that their descendants were still to be found in his own day on the royal estates

(inju) in Persia.126 We owe the unflattering descriptions of the Mongols by Amir Khusraw to the

circumstance of his being taken prisoner in 684/1285 on the defeat and death of his master Muhammad b.

Balaban.127 That the seizure of slaves continued to be an important aim cannot be in doubt. Isami's account

of the Mongol raid on the Than region suggests that the invaders fell prey to Alp Khan's forces because

they were unduly encumbered with booty and prisoners.128 Regarding other forms of plunder, the sources

have less to say. We know that the Mongols ravaged not only Muslim territories but those of the Khokhars,

whose talwaras (or talwandis) were looted and burned in 697/1298 during Keder's attack.129 Here one item

of booty may have been horses, since the Khokhar territory was among those parts of the Panjab that

produced choice mounts.130 We can also presumably take it for granted that the Mongols came in the hope of acquiring gold and silver, and in this connection we may have an explanation for India's enhanced

attractiveness. With the ambitious raids on independent Hindu kingdoms in the south from 695/1296

onwards, Delhi's rulers were known to be amassing great quantities of specie, of which the Mongols,

consequently, must have been tempted to relieve them.

126 JT, I, part 1, 189 (tr. Khetagurov, 110). For prisoners taken in the Kashmir campaign, see also

JT, III, 22 (tr. Arends, 21), and ed. Jahn, Indiengeschichte, Ar. text Taf. 61 (German tr. 56).

127

The fullest account is in DR, 36-7. See also WH, IOL ms. 412, fol. 78 (cited in Bada'uni, I,

153); and the brief allusion in DGK, 70. Mirza, Life and works, 60-2. 128 FS, 289 (tr. 464). For other references to Hindu captives, see Zafarul Islam, 'The Fatawa Firiiz

Shdhi as a source for the socio-economic history of the Sultanate period', IC 60, part 2(1986), 104 n.27.

129 KF, 33. Cf. also the statement put in Balaban's mouth in TFS, 51.

130 Ibid., 53; and see Digby, War-horse, 27-8 and n.63.

~CHAPTER 12

The military, the economy and administrative reform

The army

In 656/1258, when an expedition was being prepared to dislodge the Mongol army of Sali Noyan

from Sind, the muqta's of Kara and Awadh failed to bring their contingents.1 On other occasions, too, it is

significant that Juzjani depicts the Delhi Sultan mustering troops from 'Hindustan' or 'from the regions' (az

atraf) in order to repel Mongol attacks.2 At this stage, in other words, the forces of the Sultanate still

appear- to have been undifferentiated in terms of their respective fields of operation. Sultan Balaban is

credited with the establishment of a separate army designed specifically to combat the Mongols. It

comprised divisions under his two sons, Muhammad in Sind and Bughra Khan at Samana, of whom the

latter at least had been given the task of recruiting fresh troops, and additional forces headed by the barbeg

Begbars from Delhi: the combined total, we are told, was less than 17 or 18,000 horsemen.3 Even so, there are indications in the sources that the troops with which Jalal al-Din and 'Ala' al-Din met the invading

Mongols in the 1290s had no experience of an opponent other than the Hindus.4 This is presumably why

'Ala' al-Din is credited with efforts to recruit fresh troops with which to oppose the Mongols. The

expeditions into peninsular India were another matter, and it fell to 'Ala' al-Din, again, to raise and organize

for this purpose another force, distinct from the troops he maintained in the face of the Mongol threat.5

The evidence, slight as it is, presupposes a substantial increase in the total size of the Sultanate's

army.6 Figures for the number of troops on the sultan's muster-roll surface with regrettable infrequency in

the sources, and

1 TN, II, 76-7 (tr. 846-7).

2 Ibid., I, 471, 486, and II, 171 (tr. 667, 692-3, 1156).

3 TFS, 81; for Bughra Khan, see also 80.

4 FS, 213 (tr. 376). TFS, 257, hasham-i Hindustan.

5 Ibid., 302, 326.

6 Much of what follows is to be found in P. Jacksorr, 'Delhi: the problem of a vast military

encampment', in R. E. Frykenberg (ed.), Delhi through the ages: essays in urban history, culture and

society (Oxford and Delhi, 1986), 20-22.

~ may not be very reliable when they do.7 When Hiilegii's envoys visited Delhi in 658/1260,

Balaban intimidated them by staging a review of some 200,000 foot and 50,000 horse. How many of these

were the centre forces stationed in and around Delhi, and to what extent Balaban had drawn on levies from

the iqta's, we cannot say: Juzjani describes the troops as being brought both 'from the provinces and from

about the regions of the capital' (az atraf-u hawali-yi hadrat-i a'la).& Subsequently, Balaban as sultan was

able to review an army of 200,000 men at Awadh, on his way to crush Toghril's revolt in Bengal. Not all of

these were fighting men, however. 'Barani's characterization of them - as 'cavalry, foot, paiks, archers

[dhdnuk], pavilion-bearers [kaywani], irregulars [khwudaspa; literally 'with own horse'], archers, ghulams,

servants [chdkir], traders and bazaar people'9 - does not inspire confidence. One is reminded of Bernier's slightly contemptuous observation, apropos of Awrangzib's empire, that the numbers of the army of the

'Mogol' were inflated by the inclusion of 'servants, sutlers ... and all those individuals belonging to bazars,

or markets, who accompany the troops'.10 Nevertheless, the idea that the Sultanate was militarily strong

enjoyed a wide currency. The late thirteenth-century Maghribi geographer Ibn Sa'id thought that the

Mongols were unable to conquer India because of the numbers of men and elephants at the sultan's

disposal.11

The following- reigns appear to have- witnessed a steady expansion of the sultans' military

establishment: Word reached Mamluk Egypt of a build-up of military forces under 'Ala' al-Din Khalji.12

We can probably discount the figure of six or sevcn laks (600,000 or 700,000) of horse furnished in '

Barani's first recension for the total numbers available to 'Ala' al-Din (as also to his enemy, the Mongol

prince Qutlugh Qocha) at the time of the battle of Kili, which definitely smacks of hyperbole; according to the revised version, the sultan could raise 200,000 or 300,000 horsemen.13 Other figures come from

Mongol Persia. Wassaf and Rashid al-Din believed that the Delhi forces stood at over 300,000;14 and in his

final volume, completed some twenty or more years later, Wassaf cites 475,000 as the current size of

7 See Kolff, Naukar, 2-4, regarding the problematic figures given for Mughal armies.

8 TN, II, 83 (tr., 856, modified).

9 TFS, 86. For kaywanis, see IB, III, 415 (tr. Gibb, 753); also TFS, 400; 'Afif, 322.

10 Francois Bernier, Travels in the Mogul empire, 1656-68, ed. Archibald Constable (London,

1891), 219-20.

11 Ibn Sa'ld, Kitab al-Jughrafiyya, 163-4.

12 Ibn Abi'l-Fada'il, ed. Kortantamer, Ar. text 29, jannada'l-junud wa'l-'asakir (German tr. 109).

13 TFS1, Digby Coll. ms., fol. 96a. TFS, 267. But it should be noted that Khusraw makes Ghiyath

al-DIn Tughluq say that there were 200,000 men on the muster-rolls: Tughluq-Nama,l\.

14 Wassaf, 309; JT, ed. Jahn, Indiengeschichte, Ar. text Taf. 54, Pers. text Taf. 18 (German tr. 43);

Qashani, 183.

~the sultan's army.15 The figure of 400,000 horse gleaned by the cosmogra-pher Dimishqi (d.

727/1327), fits neatly in between.16 Under Muhammad b. Tughluq, who is said to have built up an

unprecedentedly large force within a relatively short space of time, the figures transmitted westwards, as

we shall see (pp. 260-1), are still more impressive.

The priority given to the maintenance of such vast armies entailed certain problems. In the first

place, the troops had to be kept occupied and in training. Barani describes how Balaban instituted annual

winter hunting expeditions for this purpose. They are said to have aroused Hulegu's admiration; and the

Mongol hunt, in which the game was enclosed within a vast but contracting circle (nerge), and of which

Juwayni furnishes the classic account, was indeed designed as a form of annual winter manoeuvres.17 But it

should be noted that hunts very similar to those of the Mongols - even incorporating the nerge - had been organized by the Delhi Sultans' Ghurid precursors;18 Mongol influence is therefore hardly incontestable. Be

that as it may, we occasionally glimpse Balaban's successors on large-scale hunting expeditions;19 and

under Firuz Shah, who was especially addicted to hunting, a considerable tract in Katehr was reserved for

the chase.20

The most effective means of keeping the troops in training, however, was undoubtedly regular

campaigning against the Hindus. One of Barani's complaints about the vast host raised by Muhammad b.

Tughluq for the conquest of 'Khurasan' is that it was not found possible to occupy it with holy war during

the first of the two years before it was disbanded.21 In all probability, Barani is guilty of inconsistency here,

since as we shall see part of the army was sent to Qarachil; but the remark does indicate the importance

attached to the problem of a large inactive standing army.22

Two further problems posed by the maintenance of large armies related to pay and provisions.

'Ala' al-Din determined both to maintain a standing (mustaqim) army and to do so on low pay.23 The

salaries cited - 234 tangas for a murattab, and seventy-eight tangas for a duaspa - mean little to us, since

we have no data on the level of remuneration previously available to the Sultanate's troops; and indeed the

ranks themselves are not defined. The

15 Wassaf, 528. The figure is ultimately reproduced by Firishta (I, 199-200) in connection with

'Ala' al-Din's reforms: the intermediate source is not clear, and does not appear to be Khwand-Amir.

16 Nukhbat al-Dahr fi cAja"ibi'l-Barr wa'l-Bahr, ed. A. F. Mehren, Cosmographie de Chems ed-

din Abou Abdallah Mohammed ed-Dimichqui (St Petersburg, 1866), 180.

17 TFS, 55. Hiilegu in fact died before Balaban's accession. For the Mongol hunt, see TJG, I. 19-21

(tr. Boyle, 27-9); Morgan, The Mongols, 84-5.

18 TN, I, 364-5 (tr. 385-7). Barani employs the term nerge in connection with one of 'Ala' al-Din's

hunts: TFS, 272-3.

19 E.g. Qutb al-Din Mubarak Shah: FS, 364-5 (tr. 563-4). Muhammad b. Tughluq, sometimes

accompanied by 'not more than 100,000 horsemen': MA, ed. Spies, 19 (German tr. 44)/ed. Fariq, 36 (tr.

Siddiqi and Ahmad, 44).

20 'Afif,321;and cf. 455. 21 TFS, 411. 21 See also FJ, 107. 23 TFS, 303-4.

~obscurity of Barani's account is further accentuated by the vagaries of the printed edition of the

Ta'rikh. S.H. Hodivala came nearest to elucidating what 'Ala' al-Din is actually supposed to have said,

which is: 'I shall require two horses and the corresponding equipment from a murattab, and one horse and

the equipment appropriate for one horse from a duaspa.'24 In his first recension, Barani expressly equates

the murattab with a heavy-armoured (bar-gustuwani) horseman and the duaspa with one who is not

equipped with horse-armour.25 The murattab thus emerges as a trooper who was expected to provide two

mounts, and the duaspa, paradoxically, as one whose second horse was supplied by the state. In the

circumstances, the seventy-eight tangas paid to the duaspa must reflect the lower investment in essential

war-gear which was required of him. That the murattab was the better equipped of the two emerges also

from Barani's rhetorical observation elsewhere, in the context of successful defence against the Mongols,

that one duaspa would bring in ten Mongols yoked together, while a single Muslim cavalryman (clearly the murattab is intended here) drove a hundred before him.26

In the time of Iltutmish and Balaban troopers in the centre (qalb) were paid by assignments on

villages in the districts around Delhi and in the Doab (see p. 95). 'Ala' al-DIn discarded this system. Indeed,

if a later author is to be believed, the sultan disapproved of the practice of assigning villages to ordinary

cavalrymen on the grounds that it nurtured local attachments and gave rise to regional rebellion.27 Whether

this was really the impulse behind his reform, we cannot be certain, but it would be in keeping with his

known concerns about conspiracy and revolt. At any rate, apart from a brief period under Muhammad b.

Tughluq (below, p. 262) the troops were henceforward paid in cash until the reign of Firuz Shah. The final

problem was how 'Ala' al-DIn was to pay his troops. Very early in his reign he had confiscated the property

of the vast majority of his uncle's amirs and resumed their iqta's into the khalisa. At some later point, probably in the wake of Hajji Mawla's rising in Delhi, the sultan further resumed all private property (milk)

and all existing grants, including wuquf (those to religious or charitable establishments) and in'am grants

(which were exempt from any obligation of service). BaranI says that the only people left with money were

the maliks, amirs, office-holders, Multanis and sahs (Hindu bankers and moneylenders).28

Although by

these means the sultan would have considerably augmented the resources at his disposal, we

24 Cf. with TFS, 303, the BL ms., fol. 150b: du asp-u istidad-i andaza-yi an az [here both the ms.

and the printed text insert a redundant w] murattab talbam wa-yak asp-u isti'dad-i [printed text inserts bar]

andaza-yi yak asp az duaspa [ms. has du asp in error; printed text has simply u] talbam. See further

Hodivala, Studies, I, xv and 280. The wages of each rank are again specified at TFS, 319. 25 TFS1, Digby Coll. ms., fol. 115a. The term murattab is also encountered in TN, II, 26.

26 TFS, 320. 21 cAfif, 95. 28 TFS, 283-4.

~have to assume, therefore, that much of this wealth would have had to be granted out again to

those in favour. In the interests of expanding his forces and of maintaining them on low pay, therefore,'Ala'

al-Din had recourse to other expedients. These involved (1) major changes in the taxation system and in the

collection of grain, and (2) measures designed to ensure low prices.

Taxation and the grain supply

We do not know what proportion of the crop the cultivator surrendered to the ruler prior to the

Muslim conquest of northern India: the terms employed for taxation in cash and in kind are as vague as

they are numerous.29 During the early Sultanate period, the ranas and rdutas were left to collect revenue

from their headmen (khuts, muqaddams) in order to raise the tribute to be paid to the sultan's

representatives (see pp. 99-100). The kharaj of which our sources speak was not in this period, therefore,

the Islamic land tax which it usually denotes. That tax, it has been suggested, was probably levied in the

former Ghaznawid territories in the western Panjab, and may have been extended to the immediate vicinity

of Delhi by the end of the thirteenth century; if so, our sources do not record such a development.30 'Ala' al-

DIn, however, imposed the Islamic khardj over a considerable area of northern India, setting it at 50 per

cent.31 This was the maximum allowed by the Hanafi school which was dominant in the Sultanate;32 but

'Ala' al-Din's rigour lay not so much in the percentage at which the khardj was set, but in the manner in which it was levied and in the additional taxes imposed on the cultivators.

Barani is our principal source.for 'Ala' al-Din's fiscal measures, although other writers provide odd

details and Professor Irfan Habib has argued convincingly that their cpmbined testimony on, price control,

at least, affords Barani an impressive degree, of support.33 One difficulty is that Barani refers to these

reforms twice, in two quite distinct contexts: first as part of a deliberate policy of reducing the power of

Hindu chiefs and headmen, and then several pages later, when he links the changes to the sultan's need to

make fullest use of his resources in order to maintain his unprecedentedly large army. Yet clearly the same

measures are involved. Under the new system, the revenue due was determined by means of measurement

(misdhat) on the basis of the biswa (i.e. one-twentieth of a

29 For an examination of such terms, see Gopal, Economic life, chapter 2. 30 I. Habib,'Agrarian economy', 60-1. 31 TFS, 287.

32 Nicolas P. Aghnides, Mohammedan theories of finance (New York, 1916; repr. 1969), 379-80.

Similarly, IB, IV, 223 (tr. Gibb and Beckingham, 872), says that the infidels of 'Habanq' (near Sylhet) paid

half their produce to their Muslim rulers.

33 Irfan Habib, 'The price regulations of 'Ala' uddin Khalji - a defence of Zia' Barani', IESHR 21

(1984), 394-7.

~bigha): that is to say, the yield (wafa) per biswa was estimated, and the amount due from the cultivator was arrived at by multiplying this figure by the number of biswas he held; of this total, half was

required. For the most part, the kharaj was to be paid not as a share of the crop but in cash, and Barani

alleges that the collectors demanded the tax so insistently that the peasants were compelled to sell their crop

to the grain merchants (kdrwa-niyan) as it stood in the field (bar sar-i kisht).34 One consequence of the new

system of assessment was that the contribution of the individual became all the more important. As a letter

of Ibn Mahru reveals in the middle of the fourteenth century, the peasant, though technically free (hurr asl),

was now effectively bound to the soil, since were he to abscond the total kharaj due from the village would

suffer a reduction.35

'In addition to a kharaj assessed on a new basis, 'Ala' al-Din imposed two further taxes: the chara'i, or grazing-tax, and a sukunat-ghari (or -garhi), a tax on dwellings.36 We are told little about either

impost in the standard version of the Ta'rikh-i Firuz-Shahi, but learn more about the chara i in the first

recension, where the quantity of livestock yielded (per village?) is said to comprise four oxen (sutur), two

buffaloes (gaw-i mish), two cows (mada-gaw) and twelve sheep.37 Whether the chara'i was levied on -

among others - the transhumant peoples of the eastern Panjab, whose lifestyle was predominantly pastoral,

is uncertain. We have to assume that the dwelling-tax fell on the urban population as well as on peasants. It

should be noted that unlike the kharaj such taxes enjoyed no sanction in Islamic law.

The two recensions of Barani's work differ when they come to name the official who was given

responsibility for implementing the reforms. In the first, it is Malik Yaklakhi; in the second, Sharaf al-Din

Qa'ini. Both are entitled nd'ib-wazir.38 Over a period of some years (six, according to Barani's first version),

the nd'ib-wazir saw to it that the kharaj on the basis of measurement, together with the grazing- and dwelling-taxes, were applied uniformly to a vast area, as if it were a single village. The tract in question is

defined as 'all the villages in the regions of the capital, the townships (qasabat) of the Doab country, the

land from Bhayana to Jhayin, from Palam to Deopalpur and Lahore, all the territory of Sunnam and

Samana, and from Rewari to Nagawr, from Khor to Ganuri, and from Amroha, Afghanpur and Kabar, from

Damhai to Bada'un, Kasrak and *Kotla, and the whole of Katehr'. Qa'ini is further credited with strenuous

measures to eliminate bribery and embezzlement among local

34 TFS, 287, 288, 305, 307. For the phrase wafa-yi biswa, see Hodivala, Studies, II, 97-8; Habib,

'Agrarian economy', 61. The bigha later adopted in British India approximated to five-eighths of an acre:

Yule and Burnell, Hobson-Jobson, 79 (j.v. 'beegah').

35 IM, 61-3. I. Habib, 'Economic history of the Delhi Sultanate', 297-8, and 'Agrarian economy',

54.

36 TFS, 287, 288. 37 TFS1, Digby Coll. ms., fol. 112a.

38 Ibid. TFS, 288. For the latter's nisba, see above, p. 173, n.16.

~functionaries, to the extent of inspecting the records (bahi) of the village accountant (patwari).39

A significant proportion of this region was incorporated into the khalisa, as the territories of Kol,

Baran, Mirat, Amroha, Afghanpur and Kabar and the whole of the Doab were resumed from the existing

muqta's and brought under the sultan's revenue ministry.40 Within the Doab khalisa specifically, the kharaj was to be paid entirely in grain, which was to be conveyed to the sultan's grain reserves in the capital; in the

Jhayin region, on the other hand, the tax was to be paid half in cash and half in grain, and the grain to be

stored in Jhayin and its townships.41 The enormous stores of grain kept in Delhi made a strong impression

on Ibn Battuta some decades later.42 These reserves of grain were designed for periods of famine, which

afflicted, the capital from time to time, notably in Jalal al-Din Khalji's reign, when the reserves had been

exhausted;43 though the sultan will also have had in mind the more recent crisis provoked by the Mongol

attack of 703/1303 led by Taraghai (above, pp. 222-3). We cannot dismiss the possibility, too, that 'Ala' al-

Din's vigorous campaign against the manufacture and consumption of wine and drugs, which if effective

would have entailed a loss of revenue to the state, did not spring simply from religious and moral impulses.

He may have aimed simultaneously at encouraging concentration within the agrarian sector on cereal

production.44 39 TFS, 287-9; TFS1, Digby Coll. ms., fol. 112a, adds 'and the whole of Hindustan as far as

Bengal'. Khor (BL ms. KHR) appears in error as KRH (Kara) in the printed text, which also has JHABN for

JHAYN and DBHAY for DHMHAY. Khor, an old town mentioned also at TFS, 485 (text has KHWD in error),

stands at 27° 39' N., 79° 28' E.: Hodivala, Studies, I, 296; IG, XXII, 229. For Damhai, on the route from

Bada'un to Delhi, at 28° 12' N., 78° 16' E., see Hodivala, Studies, I, 269. From the context, *K5tla (thus

TFS1; mss. of the later recension read KWYLH) must be the Kopila of Timurid sources, which Lal, Twilight,

34, identifies as Hardwar.

40 TFS 323-4. For Afghanpur, see I. Habib, Atlas, 27 and map 8A. 41 TFS, 305-6. For this interpretation of seemingly conflicting statements by Barani, see I. Habib,

'Agrarian economy', 61-2.

42 IB, III, 148 (tr. Gibb, 621): he was under the impression, however, that the stores dated from

Balaban's day.

43 FS, 217-19 (tr. 383-4). TFS, 212. For the opening of the granaries during times of dearth under

'Ala' al-Din himself, see KF, 21.

44 TFS, 284-6. Barani alleges, however, that the aim was to reduce the incidence of convivial

gatherings of the nobles which might lead to conspiracy. According to Sir George Watt, A dictionary of the economic products of India (London and Calcutta, 1889-93, 6 vols. in 9), VI, part 4, 273-4, the grapes of

the N.W. provinces and Awadh are hardly suitable for the manufacture of wine; and IB, III, 129 (tr. Gibb,

610), confirms that the grape was rare in India, being found in the Delhi region and one other province

whose name is blank in all the mss. Nevertheless, other sources suggest that wine production was

prominent in Awadh and in Kol and Mirat, all of them regions which were the object of 'Ala' al-Din's

economic reforms: TFS, 157; Mirza, Life and works, 72. It is noteworthy that at least one intoxicating

drink, bagni, was made from grain: Hodivala, Studies, I, 276.

~price control

The accumulation of stocks of grain was only partly designed as an insurance against dearth; it was also an essential component of 'Ala' al-Din's policy of price control.45 In order to maintain a large

standing army on relatively low pay, it was necessary to secure low prices of essential items. The

government therefore fixed maximum prices for a number of commodities. These comprised basic

foodstuffs - wheat (hinta), barley (jaw), rice (shall), pulse (mash, nukhud) and moth; cloth, sugar, sugar-

cane (nabat), fruit, animal fat (rawghan-i sutur) and wax (rawghan-i chiragh); and slaves, horses and

livestock.46 To oversee the maintenance of low grain prices, Malik Qabul Ulughkhani was appointed as

intendant of the market (shihna-yi manda), assisted by an intelligence officer (band), and all the merchants

(karwaniyan) were subject to his jurisdiction. The leading merchants, according to Hamid Qalandar, were

advanced money from the treasury and were paid their expenses. On the other hand, Barani says that they

had to give sureties and were obliged together with their wives and families, to take up residence in the

villages along the banks of the Yamuna. Their operations, too, were closely supervised. Hoarding and

regrating of grain - whether by cultivator, merchant or broker (baqqdl) - were forbidden, under strict penalties which included confiscation of the grain in question. In order to ensure that cultivators sold the

requisite quantities of grain to the merchants in the fields (bar sar-i kisht) and that the merchants brought it

promptly to the sultan's markets, certificates were issued by the local officials confirming that the

transaction had taken place.47 The marketing of commodities other than grain was centred on a new

institution called the Sarai-yi 'Adi, which was established in a vacant area inside the Bada'un Gate and for

which a group of prosperous Multanis were made responsible. Orders were issued that no goods were to be

sold anywhere but in the Sarai-yi 'Adl, on pain of confiscation.48

Overall responsibility for the maintenance of low prices was entrusted to a certain Ya'qub, who

combined the office of chief inspector of revenue

45 See generally Dharam Pal, 'cAla'-ud-Din's price control system', IC 18 (1944), 45-52; P. Saran,

'The economic policy and price control of Alauddin Khalji', Bharatiya Vidya 11 (1950), 195-215 (repr. in

his Studies); Shaikh Abdur Rashid, 'Price control under Alauddin Khilji', in Proceedings of the All-Pakistan

History Conference. First Session, held at Karachi ... 1951 (Karachi, [n.d.]), 203-10; I. Habib, 'Non-

agricultural production and urban economy', in Raychaudhuri and Habib, Cambridge economic history, 83,

86-7.

46 Basic foodstuffs: TFS, 305. Cloth etc.: ibid., 309, 310. Horses: ibid., 312-13. Slaves: ibid., 314.

Livestock: ibid., 315. Habib, 'Non-agricultural production', 87, points out that the price of wheat quoted by

Barani is confirmed by Hamid Qalandar, Khayr al-Majalis, ed. K. A. Nizami (Aligarh, [1959]), 185 (cf. also 241).

47 TFS, 305, 306-8. Hamid Qalandar, Khayr al-Majalis, 241, cited in I. Habib, 'Non-agricultural

production', 83. On Malik Qabul, see Hodivala, Studies, II, 100: he had presumably been a slave of the

sultan's brother Ulugh Khan.

48 TFS, 309-10. See also KF, 21-2.

~(ndzir) with those of ra'is of the capital and muhtasib of the whole empire.49 He in turn appointed

for each market an overseer (shihna) whose task was to keep prices under surveillance.50 The ra'is's

department (diwan-i riydsat) was to keep a register (daftar, tadhkira) of the names of all traders, both those

of the capital and those of the provinces. Written undertakings were required from them that they would convey agreed amounts of certain commodities annually to be sold in the Sarai-yi 'Adl.51 Twenty laks

(2,000,000) of tangas were advanced by the government to Multanis who were to convey goods from the

provinces in order to ensure cheap prices if the merchants delayed to bring their wares to the Sarai-yi

'Adl.52 For the purchase of luxury items, it was necessary to obtain a certificate (parwana) from the ra'is, in

order that traders or wealthy citizens might not buy up goods cheaply in the capital and sell them elsewhere

at a high profit.53 The * entire system rested on a network of spies, who reported abuses to the sultan.54

'Ala' al-Din's policies were reinforced by harsh penalties. To some extent the victims were

middlemen: horse-traders and horse-brokers, for instance, whose operations tended to inflate prices, were

in many cases fined or expelled from the capital and imprisoned in distant forts.55 For his part, the

uncompromising stance of the ra'is Ya'qub made him an object of terror to those who infringed the market regulations. Lashings and imprisonment were common. Flesh was cut from the faces of some offenders,

notably dealers who attempted to offset their low profits by selling short weight; they were additionally

ejected from the bazaar.56 Despite these draconian punishments, however, the government failed to

eradicate fraudulent trading.57

Purpose and effect of 'Ala' al-Din's policies

Barani is emphatic that 'Ala' al-Din's control of prices was a source of wonder to his

contemporaries;58 and indeed the policy - involving the enforcement of maximum prices for a wide range of

commodities and in some cases the elimination of middlemen - appears to have been a remarkable piece of

government interventionism, all the more impressive in the conditions that obtained during the early

fourteenth century. We should therefore be inclined to approach Barani's testimony warily, if other witnesses did not confirm that the effectiveness of 'Ala' al-Din's price control was a byword among later

generations. Ibn Battuta, visiting the Sultanate in the 1330s and early 1340s, heard 'Ala' al-Din praised in

this connection as one of the best of previous sultans, and mentions in particular

49 TFS, 317.

50 Ibid., 317-18.

51 Ibid., 309, 310-11.

52 Ibid., 309, 311. 53 Ibid., 311-12.

54

Ibid., 315, 319; see also 308 for spies in the manda.

55 Ibid., 313-14.

56 Ibid., 316, 319.

57 Ibid., 317. 58 Ibid., 305, 308, 312, 339, 340-1.

~the prices of meat, woven cloth and grain. Hamid Qalandar, writing in c. 755/1354, likewise pays

tribute to the sultan's achievement in reducing the cost of grain and to the low wages paid in his reign.59

The overall effect of 'Ala' al-Din's measures was to transfer a significantly larger share of the

agricultural surplus from the countryside to the towns and from the Hindu chiefs to the Muslim governing

class. But the essentially militaristic thrust of his economic policy is made explicit by Barani, who specifies

that the entire revenue demand (mahsul) of certain khalisa territories was set aside for the pay (wajh) of the

army (hasham) and the expenses of the imperial manufactories (karkhanas).60 He also links the control of

prices (in the first recension, the price of horses in particular) with the need to recruit soldiers on low pay.61 That the needs of the army were uppermost in the sultan's mind is also clear from the categorization of

horses into four classes, of which the lowest comprised those which would not pass muster.62 Modern

scholarly opinion has posited in addition other stimuli, though the weight ascribed to each varies. The

hypothesis advanced by Dharam Pal, who viewed the sultan's policy as also a reaction to inflationary forces

generated by the influx of gold from the south, lacks plausibility.) given that the reforms seem to have been

instituted within two or three years of 'Ala' al-Din's accession and therefore to have predated Kafur's

exploits beyond the Vindhyas.63 Shaikh Abdur Rashid believed that 'Ala' al-Din's measures were intended

to benefit not merely the state but 'the consumer at large'; and for what it is worth Hamid Qalandar does

impute, humanitarian motives to the sultan, who allegedly sought to confer benefits on his people at large.64

The most convincing analysis, however, is that of Kehrer. The drafting of peasants for the army and for 'Ala' al-Din's construction projects served to' diminish .the production of food and cloth; while the

recruitment of a certain number of foreign mercenaries would have occasioned an absolute increase in

consumption. Furthermore, in addition to the fall in supplies, there was a growing problem of distribution:

the concentration of great numbers of non-producing consumers - the troops - in the capital and its environs

accentuated difficulties in transportation from the provinces. To remedy these problems, the government

had two alternatives: controlling prices artificially and increasing the supply of money.65 Where 'Ala' al-Din

59 IB, 111,184-5 (tr. Gibb, 640-1). Hamid Qalandar, Khayr al-Majalis, 240-1.

60 TFS, 323-4. On the significance of mahsul as 'revenue demand' rather than 'produce', at least

from the time of cAfif, see Moreland, Agrarian system, 232 n.l, 249.

61 TFSf 304; cf. also 340. For horses, see TFS1, Digby Coll. ms., fol. 115a.

62 TFS, 313, anchi dar diwan nagudharad.

63 Pal, 'cAla'-ud-Din's price control system', 46. Cf. Saran, 'The economic policy', 202.

Chronological indications: TFS1, Digby Coll. ms., fol. 109a.

64 Hamid Qalandar, Khayr al-Majalis, 241. Abdur Rashid, 'Price control'.

65Kenneth C. Kehrer, 'The economic policies of Ala-ud-Din Khalji', Journal of the Panjab

University Historical Society 16 (1963), 55-66.

~had recourse to the first of these expedients, Muhammad b. Tughluq, as we shall see, would

resort to the second.

'Ala' al-Din's tax reforms subjected the khuts and muqaddams to the same assessment as the

peasants within their localities. The revenue was to be levied 'without discrimination' (bi tafawuti), so that

for this purpose the headman (khut) was treated in exactly the same way as the inhabitants of his village:

Barani says explicitly that there was to be no difference 'between the khut and the baldhar (the sweeper).66

Moreover, the chiefs' perquisites (huquq) were abolished, including their exemption from the chard'i and

ghari taxes, and it was no longer possible for them to pass their own tax burden on to those who were less well off than themselves.67 They thus suffered a twofold loss. Barani claims approvingly that the 'Hindus'

(by which he means the rural Hindu aristocracy) forfeited their surplus wealth and that their wives found it

necessary to earn wages by taking work in Muslim households. Steps were taken to reduce the potential for

rebellion. The chawdhuris, khuts and muqaddams were compelled to give up riding and bearing arms and

could no longer 'eat the betel-leaf (tanbul) - a reference to the ceremony whereby rawats (rautas) rallied to

the support of some leader, whether a Hindu prince or a rebel Muslim amir like Malik Chhajju in 689/1290

(p. 125 above). A single official of the local revenue collectorate (sarhang-i diwdn-i qasabdt), we are told,

might now rope together twenty or so of them and extract the tax from them by means of blows and kicks.

In his first recension, Barani adds that any Hindu's house in which arms were discovered became the

sultan's property.68 At a later juncture he endeavours to express the subjection of this rural aristocracy in

equally vivid terms when he depicts them guarding the highways on the sultan's behalf and keeping watch

on., caravans and travellers.69

Whether humiliation of the Hindu chiefs was the main impulse behind the reforms, however, as

Barani claims, is to be doubted: it was more probably a by-product of the government's efforts to increase

its revenue and to leave no pockets of immunity. But lest we incline to doubt the truth behind Barani's vivid

statements, 'Afif 's account of the birth of the future sultan Firuz Shah corroborates the earlier historian's

testimony regarding conditions under 'Ala' al-Din. Ghazi Malik Tughluq, at that time (c. 706/ 1306-7)

muqta' of Deopalpur, approached the local chief, Rana Mal Bhatti, and sought his daughter in marriage for

his brother Rajab. Meeting with a proud refusal, Tughluq - allegedly on the advice of 'Afif's

66 TFS, 287. For khut, see above, p. 124 n.2; for the meaning of balahar ('village menial'), I.

Habib,'Agrarian economy', 48. 67 TFS, 287, 291.1. Habib, 'Agrarian economy', 55.

68 TFS1, Digby Coll. ms, fol. 112a. TFS, 288; for the ceremony of taking up the betel-leaf, see

ibid., 182, and Hodivala, Studies, I, 265.

69 TFS, 324; cf. also 340.

~great-grandfather, who was his representative in the Abuhar district -entered Rana Mai's territory

(talwandi) and proceeded to extort the whole year's tax (mal) in cash at once, rather than in instalments (ba-

martaba) as was the usual practice. All the muqaddams and chawdhuris of the territory were beaten, and

Rana Mal's people were in great straits. When she discovered from her weeping grandmother that she was the cause of this affliction, Rana Mal's daughter told her father to surrender her to the Muslim amir and to

imagine that she had been carried off by the Mongols (why this should have afforded him any consolation

is not readily apparent). She thus became the wife of Rajab and subsequently the mother of Firuz Shah.

'Afif assures his readers that Rana Mal had no choice, for 'this was the era of 'Ala' al-Din and they were in

no position to make any murmur or outcry'.70

It is clear from this anecdote that Ghazi Malik extorted the tax direct from the headmen. Professor

Irfan Habib sees this as part of a process whereby the older rural aristocracy of ranas and rautas was

subverted. Yet at the same time a new superior rural class was emerging, and he has proposed that at its

apex stood the chawdhuri, defined by Ibn Battuta as 'the chief of the infidels' in each sadi; the sadi was a

unit of a hundred villages and doubtless corresponded to the pargana, a term first employed by 'Isami and Ibn Mahru in the middle of the fourteenth century and more commonly used by 'Afif.71

Parallel with a reduction in the perquisites of Hindu intermediaries went a growing encroachment

by the sultan's bureaucracy on the position of the Muslim muqta's. 'Ala' al-Din's expansion of the khalisa

had curtailed the area to be granted out as iqta's; though to a large extent this was compensated for by the

availability of iqta's in newly conquered territories' like Gujarat, Malwa and the Deccan. But the application

of the new method of kharaj assessment to territory which, like Awadh, was still held as iqta' would

certainly have brought about a closer supervision of the local finances by the sultan's own functionaries.72

‘Ala' al-Din's successors

According to Barani, of all 'Ala' al-Din's measures Qutb al-DIn Mubarak Shah retained only that

concerning the consumption of wine (although even

70 'Afif, 37-8. That FIruz Shah was born not in 709/1309-10 (as 'Afif, 36, claims) but in 707/ 1307-

8 is clear from cAfif s other statements that he became sultan at the age of forty-five (ibid., 20) and that he

was fourteen at Ghiyath al-Din Tughluq's accession and eighteen at that of Muhammad b. Tughluq (ibid.,

41-2); see also Hodivala, Studies, I, 390-1.

71 I. Habib, 'Agrarian economy', 56-9. For the chawdhurl, see IB, III, 388 (tr. Gibb, 741); and for

the pargana, FS, 450, 597 (tr. 680, 881); IM, 23, 146; 'Afif, 99, 236, 288, 295, 297, 339, 432, 437, 479,

483, 500; also 272 for parganadars. 72 I. Habib, 'Agrarian economy', 70.

~this was flouted).73 Price control was abandoned. Qutb al-Din proved unable to enforce it, and his

reign thus witnessed a substantial rise in the prices of grain and other foodstuffs; vendors set their own rates

for fabrics; the regulations surrounding the Sarai-yi 'Adl were discontinued, and the Multanis became

absorbed in their own commercial interests.74 The spy network fell into abeyance, and the Diwan-i Riyasat

no longer had any authority.75 Even the kharaj did not remain at the level 'Ala' al-Din had decreed,

although the extent of the reduction is uncertain. Barani asserts merely that Qutb al-DIn abolished 'the

heavy land taxes (kharajha) and burdensome requisitions from the people' and that as a consequence of the

reduction of the kharaj the 'Hindus' (again meaning, presumably, the headmen and chiefs) enjoyed ease and affluence.76 Two other pillars in the edifice constructed by 'Ala' al-Din were removed when much of the

land recently taken into the khalisa was granted out once more and the soldiers' pay, along with other

charges on the government's resources, like the stipends of the 'ulama', were increased, doubtless in

response to the rise in prices.77

Like 'Ala' al-Din, Ghiyath al-DIn Tughluq began his reign by boosting the contents of his treasury.

Although 'Ala' al-Din's grants were confirmed, the new sultan cancelled all those made by his predecessor

Khusraw Shah, and instituted, as it were, quo warranto proceedings into the rest. 'Isami, whose ancestors

thereby forfeited two villages in the Delhi region that they held by tax-free grant (in'am) from earlier

sovereigns, does not conceal his outrage at this conduct - for which, in his view, the sultan soon paid the

penalty with his life.78 Yet by Barani Tughluq's reign is depicted as one of moderation towards both the

peasantry and the amirs,. The sultan demonstrated his concern for the livelihood of ordinary peasants and for the extension of cultivation.79 The khardj was no longer to be assessed in terms of estimated yields, but

was to be based on the actual yield (hasil): the cultivators, says Barani, were thereby relieved of the

difference between the real produce and the non-existent (bud-u nabud). The amount taken as kharaj was

not to be raised by more than one-tenth or one-eleventh annually. It is accordingly clear that an increase in

the rate of the khardj was seen as desirable; but it was to be achieved in stages.80

Ghiyath al-Din Tughluq permitted the muqta's to supplement their stipends (mawajib) by retaining

up to one-fifteenth or one-twentieth of the khardj levied within their territory, as a perquisite of their

office.81 But on

73 TFS, 384: slightly later (385) he contradicts himself with the statement that not a single 'Ala'I

measure was retained.

74 Ibid., 319, 384-5.

75 Ibid.,385.

76 Ibid., 383, 385. FS, 355 (tr. 552), might suggest that Qutb al-DIn merely remitted the kharaj for

the first year of his reign.

77 TFS, 382-3. 78 Ibid., 438-9. FS, 390-1 (tr. 594-6).

79 TFS, 442.

80 Ibid., 429-30.

81 Ibid., 431, 432.

~the other hand we find the sultan warning his amirs not to encroach on the pay of their soldiers:

this shows both that a part of the revenues of the iqta' was set aside for the maintenance of the troops and

that the muqta' at this date still had access to the portion of the revenues which was earmarked for his men.82 This was to change under Muhammad b. Tughluq, whose policy further undermined the powers of

the muqta' and may well have underlain many of the rebellions of his reign.

The economy and the expansion of the Sultanate

The idea that the establishment of the Delhi Sultanate accelerated the process of urbanization over

much of northern India, as well as fostering the development of a money economy and an expansion in

craft production, is now widely accepted. Under 'Ala' al-Din, whose mint output seems to have outstripped

that of his predecessors, the increase in the kharaj and its realization in cash further contributed to the

monetization of the economy.83 We have no figures for the(tax yield from the empire as a whole prior to the

reign of the Tughluqid Sultan Firuz Shah. In the Shamsid and Ghiyathid eras the Sultanate already included flourishing ports like Lahari on the lower Indus, which Ibn Battuta was informed was worth sixty laks (i.e.

6,000,000 silver tongas) per annum to Muhammad b. Tughluq;84 there is no reason to believe that it would

have yielded, say, less than half this amount in the latter half of the thirteenth century. Although, of course,

the sultans forfeited the extensive revenue of Bengal from its secession in 685/ 1287 until its recovery by

Ghiyath al-DIn Tughluq in 724/1324, this would have been more than offset by the conquests of 'Ala' al-

Din Khalji and his successors, which greatly increased the material and fiscal resources of the Sultanate.

The fertility of Awadh and Zafarabad, over which, as we have seen (p. 200), the sultan's hold seems to have

intensified in the early fourteenth century, would be a byword at a time when the regions west of the

Yamuna were in the grip of famine during Muhammad b. Tughluq's reign.85 Ibn Battuta comments, too, on

the density of cultivation around Dhar and the prosperity of Ujjain, and on the great value of the land

revenue of the Dawlatabad province to the sultan's treasury; and indeed Muhammad fixed the revenue of

the 'Marhat' territory at six or seven krors (i.e. sixty or seventy million tangas).86 82 Ibid., 431. Irfan Habib, 'The social distribution of landed property in pre-British India', Enquiry

2, part 3 (Winter 1965), 48; see also his 'Agrarian economy', 70.

83 I. Habib, 'Economic history of the Delhi Sultanate', 289-98. H. C. Verma, Dynamics of urban

life in pre-Mughal India (New Delhi, 1986), chaps. 2, 4 and 5. Shireen Moosvi, 'Numismatic evidence and

the economic history of the Delhi Sultanate', PIHC 50. Gorakhpur 1989 (Delhi, 1990), 207-18.

84 IB, III, 112 (tr Gibb, 602)

85 Ibid., Ill, 342 (tr. Gibb, 720-1). TFS, 485, 486. 86 IB, IV, 42, 45, 49 (tr. Gibb and Beckingham, 791, 793, 795). Muhammad b. Tughluq: TFS, 501.

For the kror (= 100 laks), see Yule and Burnell, Hobson-Jobson, 276 (s.v. 'crore').

~India had long been portrayed as insatiably consuming the wealth of lands further west.87

Acquisition of the ports of Gujarat, especially, enabled the Delhi government to tap the flourishing

commerce of the Arabian Sea and the Persian Gulf; and it is perhaps no accident that in 'Ala' al-Din's time

we first encounter the malik al-tujjar, 'king of the merchants', who was responsible to the sultan for

overseeing commercial activity, or that one of al-'Umari's informants was a Karimi merchant -i.e. a member

of an important corporation of traders based in Egypt -who had twice visited Qutb al-Din Mubarak Shah.88 To the anonymous author of the Sirat-i Firuz-Shahl, Kanbhaya was 'the rendezvous of merchants, the

haven of travellers by land and by sea';89 and the affluence of its mercantile class was vividly demonstrated

in the magnificence of their mansions.90 At the beginning of the century Gujarat had attracted praise in

Rashid al-Din's history of India, and Marco Polo had earlier heard impressive tales of its manufactures.91

We know that the province produced fine cotton cloths (exported to China); was a place of transhipment for

diamonds and other precious stones; and imported black slaves from East Africa. A farman of 709/1309-10

reproduced in Amir Khus-raw's Rasa'il al-Ijaz lists numerous high-value commodities found at

Kanbhaya.92 The revenue-demand (mahsul) of Gujarat in the late 1360s is set at two krors (twenty million

tangas) by 'Afif; it is worth comparing this sum with that given for the Doab (eighty laks, i.e. eight million

tangas) at approximately the same time.93 And yet the Sultanate did not benefit merely from the possession

of outlets onto the Arabian Sea. By Muhammad b. Tughluq's reign there is fragmentary evidence that his

dominions were attracting traders from as far afield as Western Europe, who profited from the encouragement of Mongol rulers to travel by the

87 Wassaf, 300. See also JT, III, 493 (tr. Arends, 281).

88 TFS, 352. MA, ed. Spies, Ar. text 35 (German tr. 62)/ed. Fariq, 64 (tr. Siddiqi and Ahmad, 60).

The malik al-tujjar of Iltutmish's day mentioned in TN, II, 41 (tr. 790), of course, was not the Delhi Sultan's

own agent, but represented merchants from Persia, Iraq, Khwarazm and other territories outside the empire.

On the Karimis, see Gaston Wiet, 'Les marchands d'epices sous les sultans mamlouks', Cahiers d'Histoire

Egyptienne 1 (1955), 81-147; S. D. Goitein, 'The beginnings of the Karim merchants and the character of

their organization', in his Studies in Islamic history and institutions (Leiden, 1966), 351-60; M. S. Labib,

'Les marchands Karimls en Orient et sur l'Ocean Indien', in M. Mollat du Jourdain (ed.), Societes et compagnies de commerce en Orient et dans l' Ocean Indien (Paris, 1970), 209-14.

89 SFS, 21, marja'-i tujjar-u ma'man-i suffar (tr. Basu, in JBORS 23 [1937], 99). See also the

author's reaction to Kanbhaya in IM, 133.

90 IB, IV, 53, 55 (tr. Gibb and Beckingham, 797, 798).

91 JT, ed. Jahn, Indiengeschichte, Pers. text Taf. 13, Ar. text Taf. 51 (German tr. 36). Marco Polo,

tr. Moule and Pelliot, I, 420-1, 422-3/tr. Yule and Cordier, II, 393-4, 398-9.

92 RI, IV, 141-3. See generally Simon Digby, 'The maritime trade of India', in Raychaudhuri and

Habib, Cambridge economic history, 139-40, 142, 149; V. K. Jain, Trade and traders in Western India (AD 1000-1300) (New Delhi, 1990), 98-105.

93 'Afif, 221,296.

~overland route through Urgench and Ghazna, for a group of Venetians is known to have visited

Delhi in 1338.94

Changing priorities

Alongside the marked increase in the revenue from conquered territory, however, the sultans'

government relied on the fruits of predatory campaigns against the Hindu powers of the subcontinent. The Mongol threat appears to have modified the order of priorities within the framework of military policy, for

as early as 645/1247 Juzjani has Ulugh Khan Balaban advocate the looting of Hindu territory not merely in

order to chastise the infidel but to amass booty which could then be used to maintain a defensive army in

the face of Mongol invasions.95

The fact that Juzjani wrote as a contemporary, and still more his proximity

to Ulugh Khan, make it very likely that these sentiments illustrate the adoption of a conscious policy by

Delhi's rulers following the intensification of Mongol pressure after 1241. They contrast sharply with the

more simplistic analysis of Barani, who depicts Balaban (now sultan) as refusing to launch campaigns

against the Hindus as long as the Mongol menace persisted.96 This statement is in any case rendered suspect

by the passage that follows: a ringing denunciation of the expansionist policy, leading as it does to the

overtaxing of resources and possibly rebellion, with the ultimate consequences of bloodshed and harsh punishments.97 Clearly what Barani had in mind here was not Balaban's reign at all, but the recent chaos

caused by the expansionist designs of Muhammad b. Tughluq.

A certain degree of military activity against the Hindus was vital both to keep the armed forces in

proper training and also to harvest the resources with which to reward them; otherwise it would have been

far more difficult to maintain a large army to repel the Mongols. Hence we have good reason to distrust

Barani again when he describes how Taraghai's invasion prompted 'Ala' al-Din Khalji to give up

'campaigning and taking fortresses' (lashgarkashi-u hisargiri).98 We know in any case that this was simply

not so: even were we to disregard the expeditions which 'Ala' al-Din personally headed against Siwana and

Jalor, and which Barani fails to mention, the notice he gives of Malik Na'ib Kafur's campaigns in the south

would alone indicate that the above statement is worthless. It might have been interpreted to mean that in

the face of Mongol pressure 'Ala' al-Din confined himself to plundering raids and abandoned the policy of outright annexation instanced in the fate of Ranthanbor and Chitor; but this inference is precluded, again,

by the annexation of Malwa from 705/1305 onwards and

94 R. S. Lopez, 'European merchants in the medieval Indies: the evidence of commercial

documents', Journal of Economic History 3 (1943), 174-80.

95 TN, II, 57 (tr. 816).

96 TFS, 50-1.

97 Ibid., 51-2, 53. 98 Ibid., 302.

~of Deogir in c. 1314. Nevertheless, under 'Ala' al-Din the two arms of military policy -

plundering operations and the imposition of direct rule -appear at least to have been kept in tension. The

succeeding reigns, by contrast, witnessed a steady move towards the absorption of vast tracts of territory

into the empire.

~CHAPTER 13

Stupor mundi: the reign of Muhammad b. Tughluq

The reign of Muhammad b. Tughluq throws up perhaps more problems than any other in the

history of the Sultanate. At the sultan's accession the authority of Delhi was acknowledged over a larger

area of the subcontinent than under any previous monarch. It is to this process of expansion that Barani

refers when he describes the unprecedented scope and efficiency of the revenue department in

Muhammad's early years.1 And yet the reign appears to be dominated by an extraordinary number of

revolts. By the sultan's death in 752/1351 Bengal and every tract south of the Vindhyas had declared their

independence, and none of these provinces was ever recovered.

In the revised version of his Ta'rikh, Barani blames the disasters of the reign on the sultan's

chimerical designs.2 But it needs to be borne in mind that by the accession of Muhammad b. Tughluq, a

policy of direct rule was progressively replacing that of plundering and levying tribute on Hindu kingdoms. The absorption of such vast areas of territory brought its own problems in its wake; and they were very

probably a major factor underlying the acute economic difficulties which overwhelmed the Sultanate in the

1330s. Launching regular attacks on enemy territory in order largely to finance a sizeable standing army for

other purposes was one matter; it was quite another to maintain garrisons and a civil administration in a

conquered province, with all the expense involved in annual accounting and transportation of revenues.3

Newly acquired provinces, moreover, could not be treated in the same rapacious manner that characterizes

warfare in enemy country. The Delhi Sultans therefore suffered a twofold loss. It would have been most

keenly felt, perhaps, in respect of gold bullion, which had loomed so large in the looting campaigns of 'Ala'

al-Din and Kafur. The problems were exacerbated by Muhammad b. Tughluq's extraordinary expenditure

and proverbial generosity.4

Wider economic trends, too, about which we are imperfectly informed,

1 TFS, 468-9.

2 Ibid., 471.

3 As Barani in fact realized: ibid., 51-2.

4 Firishta, I, 239, says that he was spending the treasure amassed by 'Ala' al-Din Khalji.

~may have contributed to the problems during Muhammad's reign. It is easily forgotten that two

great Mongol powers - the Ilkhanate and the Chaghadayid polity in Central Asia - also underwent considerable upheavals during the second quarter of the fourteenth century, as did the Golden Horde

slightly later,5 and that the Mamluk Sultanate was a prey to monetary crises during the third reign of Sultan

al-Nasir Muhammad b. Qala'un (709-741/1310-1341).6 This suggests that the Delhi Sultanate and its

neighbours and major trading-partners may have been enveloped in a common economic turbulence; but

firm conclusions must await further research.

Opposition from Tughluq's old adherents

The sultan's initial attempts to intensify his authority in the provinces seem to have lain behind

three insurrections during the years 727-8/1326-8. On the face of it, the revolts of his cousin Baha' al-Din

Garshasp, at Sagar in the Deccan, and of Kushlii Khan in Sind are puzzling. Both men had played a central role in the revolt of Tughluq against Nasir al-DIn Khusraw Shah in 720/1320 and were among the many

adherents of the Tughluqid regime who were confirmed in office at Muhammad's accession. It looks as if

Muhammad's own policies may have alienated these leading amirs whom he had inherited from his father.

Tughluq had banned informers (munhiydn) from the iqta's, but Ibn Battuta tells us that Muhammad

employed a network of spies who reported his amirs' actions to the sultan;7 at what stage the practice had

been reintroduced, however, we cannot be certain. More importantly, Barani's statements that during the

first few years of the reign the accounts even of far distant provinces were audited on just the same basis as

were those of the Doab, and that a hundred or two hundred orders arrived daily in the office of the

kharitadar for transmission to the walis and muqta's,8 are a sign that the new sultan was from the outset

exercising a far closer supervision over the affairs of the provinces than his predecessors had done. The

great provincial governors, for whom Ghiyath al-DIn Tughluq had been primus inter pares, must have

received the distinct impression that his son aimed to preside over a centralized despotism.

More particularly, the creation of a second capital at Dawlatabad (formerly Deogir) in the Deccan

may have played its own part in prompting

5 Boyle, 'Dynastic and political history', 413-16. Barthold, Zwolf Vorlesungen, 205-9, and Four

studies, I, 134-8. P. Jackson, 'Chaghatayid dynasty', Enc.Ir., V, 346. Berthold Spuler, Die Goldene Horde:

die Mongolen in Rutland 1223-1502, 2nd edn (Wiesbaden, 1965), 109ff.

6 Hassanein Rabie, The financial system of Egypt A.H. 564-7411 A.D. 1169-1341 (Oxford and

London, 1972), 189-97. Jere Bacharach, 'Monetary movements in medieval Egypt, 1171-1517', in J. F.

Richards (ed.), Precious metals in the later medieval and early modern worlds (Durham, NC, 1984), 167.

7 TFS, 429, for Tughluq. IB, III, 343-4 (tr. Gibb, 721), for Muhammad's spies.

8 TFS, 470.

~these two insurrections. About the affair at Sagar we are told very little; but in view of its

proximity to Dawlatabad, Garshasp conceivably felt threatened by the establishment there of a new bastion

of central power. We know more about Kiishlu Khan's rising. According to Sirhindi, the sultan sent an

officer to superintend the removal of Kushlu Khan's family and household to court (i.e. to Dawlatabad), and

the officer's arrogant behaviour stung the amir's son-in-law into murdering him.9 Both insurrections were crushed. Garshasp was defeated by Ahmad b. Ayaz, taking refuge first with the Hindu ruler of Kampila and

then with the Hoysala king Vira Ballala III, who handed him over to the sultan's forces for execution.

Muhammad personally moved against Kiishlii Khan, who was defeated and killed.10

The third rising, that of Ghiyath al-Din Bahadur Bura in Bengal, seems to represent nothing more

than a bid by the previously sovereign dynasty to throw off Tughluqid overlordship. At his accession,

Muhammad had released Bura from prison, conferred a chatr on him, and sent him to Sunarga'un, where he

was to enjoy the status of joint ruler with Muhammad under the watchful eye of the sultan's adopted brother

Bahram Khan, from Lakhnawti. Bura revolted - probably in or after 728/1327-8, when his coins still carry

Muhammad's name - but was overthrown by Bahram Khan with the aid of reinforcements from Delhi.11

The fate of Bura's brother Nasir al-DIn is unknown: there may be some connection with an attempt by 'the

amirs and grandees of Lakhnawti' who were with the sultan in Delhi at the time of Tarmashirin's invasion (c. 729/1328-9) to return to their own country and stir up rebellion.12 But in any event it appears that for the

next few years Bengal was administered by officers appointed by the Delhi Sultan.

9 TMS, 99-100.

10 Garshasp: TFS1 Digby Coll. ms., fol. 161a (Bodleian ms, fol. 192b, has KYTHL in error for

KNPL); FS, 424-31 (tr. 651-9); IB, III, 318-21 (tr. Gibb, 710-11). Kushlu Khan: TFS, 478-9; FS, 435-43 (tr.

663-72); IB, III, 321-4 (tr. Gibb, 711-12), ascribing the rift with the sultan to Kiishlii Khan's refusal to

exhibit the skins of Garshasp and Bahadur Bura; but see below. The date of Garshasp's rebellion is known

from an inscription of Nov. 1326: P. B. Desai, 'Kalyana inscription of Sultan Muhammad, Saka 1248', El

32 (1957-8), 165-70. Kiishlu Khan's revolt is dated in 'the latter part of that year' [727] in TFS1, Bodleian ms., fol. 191b/Digby Coll. ms., fol. 160a. The Lahore campaign in Jumada II 728/April 1328 (Siyar, 215)

must have been part of Muhammad's operations against Kushlu Khan.

11 IB, III, 316-17 (tr. Gibb, 709-10); also IV, 213 (tr. Gibb and Beckingham, 869). He erroneously

makes it out to be anterior to Kiishlii Khan's revolt, however, since he has the skins of Garshasp and Bura

circulated round the empire at the same time. FS, 422, 444 (tr. 648, 673), is brief. For coins of Bahadur

Bura, see CMSD, 130 (no. 505C). I. Prasad (Qaraunah Turks, 150), Husain (Tughluq dynasty, 223) and

Nizami (in HN, 506) all date his revolt in 730/1329-30.

12 Only in TFS1, Bodleian ms., fol. 192a/Digby Coll. ms., fol. 161a.

~The creation of a second capital at Dawlatabad and military build-up at Delhi

In the first recension of his work, Barani, who in the standard version supplies no date for the so-

called transfer of the capital to Dawlatabad, places it in 727/1326-7.13 The abandonment of the project can

be dated to the time of Muhammad's visit to Dawlatabad in c. 736/1335-6, on his way back from the

abortive expedition to suppress the rebellion in Ma'bar, when we are told that he granted permission to

those who wished to return to Delhi.14 Although the element of compulsion cannot be denied and

conditions on the journey to Dawlatabad were surely difficult, even 'Isami alludes in passing to the fact that

those citizens of Delhi who cooperated received gold from the treasury. Barani amplifies this by stating that

the shaykhs, 'ulama' and riotables of the capital were allotted cash and villages in the Deogir territory, and

that the government purchased from the ordinary citizens their houses in Delhi.15 Arbitrary the project may

have been; but its enforcement was not conducted in a totally unfeeling manner.

Who was required to move south, however, and to what extent Delhi was left deserted, have been

a matter of dispute. In his first recension Barani depicts two stages, of which the former comprised the

transfer of the sultan's mother, Makhduma-yi Jahan, and her household, together with those of the

grandees; the latter exodus, following on Kushlii Khan's revolt (and therefore to be placed in or after

728/1327-8), involved the people of the townships (qasabdt) around the capital as well as those of Delhi.16

But claims in the sources that the city was completely emptied of its inhabitants are deeply suspect. Husain

cites the testimony of Sanskrit inscriptions of 1327-8 indicating that Hindus continued to live in the vicinity

of the old capital;17

and Sirhindi refers to the 'vulgar and riff-raff (mardum-i 'awamm-u awbash) left behind

to plunder the goods of the citizens.18 In his first recension, Barani says that the sultan had the 'ulama' and shaykhs of the 'districts and townships' (khitat-u qasabdt) brought to live in the city and given pensions and

stipends.19

In order to understand Muhammad's so-called transfer of capital, it is necessary to recognize that

for our sources 'the people' (khalq) denoted the

13 Ibid., Bodleian ms., fol. 190b/Digby Coll. ms, fol. 159b.

14 TFS, 481. For the date of Muhammad's departure on the Ma'bar campaign (9 Jumada I 735/5

Jan. 1335), see IB, III, 427, in conjunction with Venkata Ramanayya, 'Date of the rebellions', 141 and n.l

(correcting the year 1341 given in Gibb's tr., 758).

15 FS, 446 (tr. 676). TFS1, Bodleian ms., fols. 191b, 192a/Digby Coll. ms., fol. 160. Cf. also TMS,

102.

16 TFS1, Bodleian ms., fols. 191, 192a/Digby Coll. ms., fol. 160. TMS, 102, speaks of the

inhabitants of Delhi and the qasabat.

17 Husain, Tughluq dynasty, 146-8: these epigraphs are now most readily accessible in P. Prasad,

Sanskrit inscriptions, 22-31 (nos. 1:10 and 1:11). Roy, 'Transfer of capital'. 170-1, however, dismissed the

inscriptions as irrelevant.

18 TMS, 102. 19 TFS1, Bodleian ms., fol. 192a/Digby Coll. ms., fol. 160b.

Map 5: The cities of Delhi

~more illustrious Muslim families of the capital.20 And even the term shahr, 'the city', when

employed in the context of Delhi, is susceptible of two meanings. When Barani talks of 'the city', he

sometimes means simply the old city of Delhi - Qil'a Rai Pithura, the city of Aybeg and Iltutmish - as

opposed to the entire complex of settlements and royal residences -Kilokhri, Siri, Hazar Sutun and

Tughluqabad - that had grown up in the intervening decades.21 During the very time that he is known to

have been transferring personnel from Delhi to the Deccan, Muhammad was engaged in ambitious new

construction projects within the Delhi region. He built in 727/1326-7 a new fortress, 'Adilabad, not far from

Tughluqabad, and linked the old city of Delhi to Siri with walls that enclosed an area henceforth known as

Jahanpanah.22 It is evident from Ibn Battuta's 20 As observed by Husain, Tughluq dynasty, 149, 152 n.2.

21 E.g., TFS, 449-50.

22 Hilary Waddington, 'Adilabad: a part of the "fourth" Delhi', Ancient India 1 (1946), 60-76.

Anthony Welch and Howard Crane, 'The Tughluqs: master builders of the Delhi Sultanate', Muqarnas 1

(1983), 128-9. For the date 727, see Husain, Tughluq dynasty, 167 and n.2, citing Badr-i Chach.

~account that Muhammad, like his Khalji predecessors, resided in the palace of Hazar Sutun,

which had been built by 'Ala' al-Din outside Siri and lay within Jahanpanah. According to the same author, the sultan had intended at one point to surround all four 'cities' (old Delhi, Siri, Jahanpanah and

Tughluqabad) with a single wall, but relinquished the idea in view of the expense involved.23

It is hard to reconcile this extensive programme with the notion that Muhammad envisaged the

abandonment of the entire Delhi conurbation. What really seems to have occurred is that the principal

Muslim residents of the old city, with their large households, were despatched to Dawlatabad.24 Excepted

were the military. The exodus is known to have included the households of the grandees and provincial

governors; but during the two years in which the sultan remained in Delhi following the suppression of

Kushlu Khan's insurrection, says Barani, 'the amirs, maliks and troops' were with him, while their families

were in Dawlatabad.25 The old city was not deserted, precisely because it was being turned into a military encampment, a development closely connected with the recruitment of an enormous army for Muhammad's

so-called 'Khurasan project' which will be discussed later. This is surely what 'Isami is referring to when he

asserts that the city was repopulated with those whom he scornfully terms 'rustics' from the surrounding

territory (parganat) and who were clearly Hindus.26 The two projects - the invasion of Khurasan and the

partial emigration to Dawlatabad - had to coincide, as one source indicates they did,27 in order to minimize

the increase in consumption in Delhi and the setting of impossible targets for the grain producers. Nor does

it appear that the sultan had miscalculated here, since Barani ascribes the disbandment of the Khurasan

force after one year not to a shortage of supplies but to a dearth of funds to pay the troops.28

Regarding the size of the Khurasan force, BaranI supplies conflicting details. In his first recension,

he cites a figure of 470,000 on the testimony of the na'ib-i 'arid himself, Zahir al-Juyush; the later version

tones this down to 370,000 and does not mention his informant.29 This has been taken as the total number of men in the sultan's army, which recalls the comparable figures for 'Ala' al-Din's reign and thus renders

Muhammad's Khurasan force much less remarkable.30 But whichever number we choose to accept, these

figures clearly apply, rather, to a specially raised force, over and above the usual total for the military

establishment.31 Al-'Umari was told that

23 For Hazar Sutun, see IB, III, 220, 399 (tr. Gibb, 660, 746); for the wall, ibid., Ill, 147 (tr. Gibb,

619, 621).

24 TFS 473 khawass-i khalq ... mardum-i guzida wa-chida. See also Husain, Rise and fall. 11

Off.; Tughluq dynasty, 146ff.

25 TFS, 479.

26 FS, 450, 453 (tr. 680-1,684-5).

27 Siyar, 111.

28 TFS1, Bodleian ms., fol. 201b/Digby Coll. ms., fol. 167a. TFS, 477.

29 TFS1, Bodleian ms., fol. 201b/Digby Coll. ms., fol. 167. TFS, 477.

30 Digby, War-horse, 24 and n.41a.

31 This is the testimony at least of Firishta, I, 240.

~Muhammad's troops in the capital and in the provinces totalled 900,000. Al-Safadi, however,

who reproduces this figure on the authority of an official envoy from Delhi to the Egyptian Sultan al-Nasir

Muhammad, 'Abd-Allah 'Daftar-khwan', is sceptical, adding that the true number is reputed to be nearer

600,000.32

That the mustering of such a vast army posed difficulties for the government was due particularly,

it seems, to a change in the system of remuneration. As al-'Umari was told, all the troops now received pay

from the sultan's (revenue) ministry (diwan).33 It was also during these years that Muhammad introduced

the token currency - actually a low-denomination bronze (muhr-i mis) coinage - which Barani again links implicitly with the recruitment of large numbers of troops and Sirhindi with the need for cash advances to

Delhi's new inhabitants.34 The plan has also to be viewed against the background of the quickening pace of

commerce (pp. 252-3 above) and of pressure on the gold-silver parity of 10:1 that underpinned the

monetary system. The dethesaurization of large quantities of gold seems to have upset this ratio,

accentuating the problem of a shortage of silver that had grown more acute by Muhammad's reign.

Indications are seen in his earlier issues of debased silver tangas since 727/1326-7 and in the urgency with

which Qadr Khan, his governor in Bengal (a region which through commerce enjoyed access to plentiful

supplies of silver in Yiin-nan and Burma), would amass large quantities of coined silver for despatch to

Delhi prior to his assassination and the rebellion of the province in c. 1361 1335-6.35

The Qarachil

expedition and Muhammad's attack on Nagarkot in 738/1337 were doubtless also actuated by a need for silver. That the gold-silver ratio had temporarily worsened is shown by the remark, in a geographical work

composed in Persia in c. 740/1339, that Muhammad had terminated the practice of hoarding treasure and

was spending his gold reserves. His heavy expenditure had caused a fall in the price of gold, so

32 MA, ed. Spies, 12-13 (tr. 37)/ed. Fariq, 24 (tr. Siddiqi and Ahmad, 37). Al-Safadi, Wafi, III, 173

(tr. Khan, 187); but cf. his A'yan al-'Asr, fol. 3a, which reads 700,000' for 900,000: his informant may have

been the hajib 'Abd-AlIah who arrived in Persia as Muhammad's ambassador in 1327-8: Shabankara'i, 288.

Ibn Hajar al-'Asqalani (d. 852/1449), al-Durar al-Kaminafi A'yani'l-Mi'ati' l-Thamina (Hyderabad, Deccan,

1348-50/1929-32, 4 vols.), Ill, 461, follows Wafi but cites only the lower number of 600,000. The figure of

forty laks (four million) given for the infantry, lastly, by Ibn Abi'l-Fada'il, ed. Kortantamer, Ar. text 27

(German tr. 104), is doubtless due to a confusion of units and should perhaps stand at 400,000; his figure

for cavalry is 300,000. 33 MA, ed. Spies, 13 (German tr. 37, 38)/ed. Fariq, 24, 25 (tr. Siddiqi and Ahmad, 37-8).

34 TFS, 475; also TFS1, Bodleian ms., fol. 201b/Digby Coll. ms., fol. 167b, where it is linked with

the sultan's generous gifts as well as the raising of troops. TMS, 102.

35 Simon Digby, 'The currency system', in Raychaudhuri and Habib (eds.), Cambridge economic

history, 97-8, especially the quotation from TMS, 104-5. Moosvi, 'Numismatic evidence,' 215-16. CMSD,

162-6. For sources of silver, see John Deyell, 'The China connection: problems of silver supply in medieval

Bengal', in Richards (ed.), Precious metals, 207-27.

~that it was no longer economical to export it to India and the direction of this traffic was now

reversed.36

We have to ignore most of the somewhat jejune account given by 'Isami, who refers to coins made

of iron and leather as well as bronze; though his assertion that the coins were current over a period of three

years is corroborated by those relatively numerous pieces that have survived, which bear dates from 730 to

732.37 Barani's fuller narrative suggests that the scheme failed owing to widespread forgery of the coins in

the countryside by Hindu chiefs and their agents, who accordingly used them to pay the land-tax. In this

fashion great quantities of bronze coins reached the treasury, giving rise to a loss of confidence and a

depreciation of their value; the government was obliged to recall the coins and to issue gold and silver

tangas in exchange.38 But Baranl's frequent claims that the treasury was emptied as a result of the project

(or indeed Muhammad's other policies) must be treated with caution. Had this been so, Muhammad would have been in no position to redeem the bronze coins; still less would he have been able to advance huge

sums to the peasantry for the purpose of restoring cultivation (see below).

It is nevertheless a measure of the strain placed on the sultan's finances by the Khurasan project

that 'Ala' al-Din's system was abandoned, the army being paid partly in cash and partly in iqta's,39 and that

in order to pay his considerably increased army Muhammad imposed on his subjects in the Doab a heavier

burden of taxation than even 'Ala' al-Din had done. Any increase in taxation, following so swiftly on

Tarmashirin's devastation of the province (see above, p. 232),40 would have provoked severe discontent; but

the precise nature of the measure is unclear. Baranl's claim in the standard version of his Ta'rikh that the

kharaj underwent a ten- or twentyfold (yaki ba-dah wa-yaki ba-bist) increase was rightly dismissed by

Moreland as a mere rhetorical device.41 The kharaj, as far as we know, already stood at the fifty per cent established by 'Ala' al-Din, the legal maximum according to the Hanafi school. But there are further hints.

Firstly, Barani suggests in his earlier recension (which is even vaguer regarding the proportion of the

increase) that what the peasants found so intolerable was that they were now being required to pay at least a

part of the assessment in cash (zar, 'gold'); and in the second place he refers to other numerous and heavy

exactions (abwab).42 The uncanonical taxes

36 Hamd-Allah Mustawfi Qazwini, Nuzhat al-Qulub, ed. and tr. Guy Le Strange, The geographical

part of the Nuzhat al-Qulub, GMS, XXIII (Leiden and London, 1915-19, 2 vols.), I (text), 230, an zarhara

sarf mikunad, and II (tr.), 222.

37 FS, 459-61 (tr. 693-5). CMSD, 139-46 (nos. 574-616).

38 TFS, 475-6; TFS1, Bodleian ms., fols. 201b-202b/Digby Coll. ms., fols. 167b-168a.

39 TFS, 476-7.

40 Referred to explicitly in this context by TMS, 113.

41 TFS, 473. Moreland, 48 n.l; also I. Prasad, Qaraunah Turks, 71-3. TMS, 101-2, has yaki ba-blst,

but later (113) says one in ten or one in twenty. 42 TFS, 479, shada'id-i mutdlaba wa-bisyari-yi abwab; see also 473. TFS1, Bodleian ms., fol.

192b/Digby Coll. ms., fol. 161a.

~abolished by Muhammad's successor included a whole range of imposts, over and above the

house tax (gharl) and grazing tax (chara i) which had been instituted by 'Ala' al-Din (above, p. 243).43 It is

accordingly possible that many of these imposts were innovations dating from Muhammad's reign and that

the phrasing in Barani's later recension is meant to signify a considerable increase in the total number of

taxes levied rather than in the percentage of income taken by the state. Lastly, Sirhindi speaks of all three

taxes being levied with much greater rigour: the yield assessed was a standard one rather than the actual

harvest, and the value was calculated according to decreed prices and not those current in the market.44

The Khurasan project and relations with the Mongols

The object of Muhammad's heavy expenditure on the military in the years from 1329 onwards was

the taking of the offensive against the Mongols. There had long been a tendency for the sultans to look over

their shoulder at the prospect of expansion beyond the Indus, inclinations which were doubtless encouraged

by the numerous refugees from these regions at their court. The spectacular success of the Delhi Sultanate

in reducing and governing an unprecedentedly large proportion of peninsular India may well have

furnished a fresh inducement for Muhammad in particular to turn his attention to the north-west. We have

seen (p. 231) how at the very beginning of his reign he headed an expedition to the Mongol frontiers;

although Tarmashirin's invasion seems to have been the immediate impulse behind the 'Khurasan project'.

Barani's misleading use, at one point in the standard recension, of the phrase 'Khurasan and Iraq'

for the territories that were the object of Muhammad's designs45 has needlessly confused the issue. The

term 'Khurasan' is itself ambiguous. For the inhabitants of India during the Sultanate period, and even as late as Babur's era, it denoted loosely the territories west of the Indus.46 Ishwari Prasad and Agha Mahdi

Husain therefore concluded that Muhammad planned to attack the Ilkhanate.47 But 'Khurasan' also

designated the regions that today comprise northern Afghanistan and were at this time subject to the

Chaghadayid khans. In his first recension Barani is more specific, referring to the object of Muhammad's

ambitions as the 'upper country' (aqalim-i bala or bala-dast); and at one point in the later version he, in

common with other sources, speaks of

43 FFS, 5 (tr. Roy, 453).

44 TMS, 101-2: he was under the false impression that the chara' i and the gharl had been

introduced by Muhammad. I. Habib, 'Agrarian economy', 63. 45 TFS, 476.

46

IB, III, 229 (tr. Gibb, 664). Babur-Nama, I, 202.

47 I. Prasad, Qaraunah Turks, 118-24; Husain, Tughluq dynasty, 138-43. But cf. Aziz Ahmad,

'Mongol pressure', 189, though he erroneously dates the Khurasan project after Tarmashirin's death.

~the sultan's plan to conquer 'Khurasan and Transoxiana (Ma wara' al-Nahr)'.48

From this we can

be certain that Muhammad intended to attack the old enemy, the Chaghadayids; with the Ilkhanate his relations were in fact amicable (above, p. 233). Professor Siddiqui has suggested that one reason for the

abandonment of Muhammad's plans was the onset of friendly relations between the sultan and Tarmashirin,

although Barani, intent on surveying the sultan's internal policies, makes no mention of this.

Barani says that part of the Khurasan force was sent to Qarachil.49 Like 'Khurasan', this is a highly

unspecific term, which in its broadest sense denotes the entire Himalayan range.50 But it is clear that in the

particular context of Muhammad's ambitions the sources are referring to a major Hindu principality. In an

article published some years ago, I proposed that the objective of the sultan's army was Kashmir, which is

known from indigenous sources to have undergone at least two invasions during the second quarter of the

fourteenth century.51 There are admittedly difficulties with this identification, but Muhammad allegedly

envisaged sending a Muslim divine to Kashmir around this very time,52 and some tract in the north-west

must be in question, given the connection with the Khurasan project which Barani makes so emphatically:

It occurred to Sultan Muhammad that since the preliminaries (pish-nihadha) for the conquest of

Khurasan and Transoxiana had been effected (dar kar shuda ast), the Qarachil mountains, which lay on the

direct route (dar rah-i nazdik), as a boundary and a screen between the empire of India and the empire of

China, should be subjected to the banner of Islam, so that the path of the army's advance and the entry of

horses should be made easy.53

It is to be noted that the mention of China, which misled the seventeenth-century compilator

Firishta into believing that Muhammad planned the conquest of that country,54 is purely incidental. From

Barani's phrasing, it looks as if one purpose was to protect the route by which bala-dasti war-horses entered

the Sultanate. It is thus hard to see how the sultan would have been interested, for example, in sending part of the Khurasan force into the Kumaon-Gahrwal region.55 Whatever the case, the Delhi forces were lured

into the mountains and there annihilated by the enemy; only a fraction of the army returned. Sirhindi sets

the total strength of the force at 80,000 horse, excluding servants (chakir) and slaves; 'Isami gives one lak

48 TFS, 477. TFS1 Bodleian ms., fol. 200a/Digby Coll., fol. 166b, reads 'Khurasan, 'Iraq and Ma

wara' al-Nahr', but cf. fol. 201b/fol. 167a, aqalim-i bala-dast. Siyar, 271 ('Khurasan and Turkistan').

Siddiqui, 'Sultan Muhammad bin Tughluq's foreign policy', 9-10.

49 TFS, 477.

50 E.g. IB, III, 325, 438-9 (tr. Gibb, 713, 763).

51 Jackson, 'The Mongols and the Delhi Sultanate', 135-42; but cf. Siddiqui, 'Sultan Muhammad

bin Tughluq's foreign policy', 15 and n.45.

52 Siyar, 228.

53 TFS, 477 (to be corrected from BL ms., fol. 236b).

54 Firishta, I, 240. I. Prasad, Qaraunah Turks, 126-8, 134-6, was rightly sceptical.

55 As proposed by Prasad, ibid., 128-31, and by Nizami (HN, 522), who (misquoting Barani)

dismisses any connection between the Khurasan project and the Qarachil enterprise.

~(100,000) of horse, of whom 5-6000 returned.56 The figure given by Barani in the course of a

very brief account in his first recension is significantly lower, at 30,000 or 40,000; his assertion in the

revised text that the survivors totalled a mere ten horsemen is an obvious hyperbole.57

Subsequently,

alleges Ibn Battuta, Muhammad was able to come to terms with the inhabitants of Qarachil, who undertook

to pay him tribute: that they had become tributary to Delhi is further confirmed by al-'Umari, though he

does not mention the sultan's failure to overcome them by military means.58

Peasant revolt and economic dislocation

Muhammad's enhanced revenue demands provoked a widespread revolt among the cultivators in

the Doab, who burned their crops, drove off their cattle and took refuge in the jungles. Having first ordered

his revenue officers (shiqqddrdn) and military commanders (fawjdaran) to plunder the recalcitrant

territories, the sultan subsequently took the field in person and mounted punitive attacks on Baran and Kol.

The uprising probably occurred in c. 1332-3, but it appears that Muhammad headed two expeditions into

'Hindustan' and that his operations in the vicinity of Qinnawj and Dalmaw (where he was absent at the time

of Ibn Battuta's arrival in Delhi in 734/1334) likewise formed part of his attempt to suppress the Doab

rebellion.59 The failure of grain to reach Delhi from the Doab gave rise to famine, and the situation was

exacerbated by the onset of a lengthy period of drought following the sultan's return from his Ma'bar

expedition.60 Barani speaks of its impact on Delhi, many of whose inhabitants either perished or fled into

the countryside; and it is surely to this date (sc. 735-6/ 1335-6) that we must ascribe the comment by Ibn

Battuta that he found the capital relatively deserted.61 It appears, however, that a far wider area came to be affected by famine, for when Muhammad had passed through Malwa en route for Ma'bar, he had found the

network of runners (the dhawa) along the route abandoned, and similarly we read of famine in the town-

56 TMS, 114. FS, 467 (tr. 703).

57 TFS1, Bodleian ms, fol. 193a/Digby Coll. ms, fol. 161b. TFS, 477-8.

58 IB, III, 327-8 (tr. Gibb, 714). MA, ed. Spies, 5 (German tr. 23)/ed. Fariq, 11 (tr. Siddiqi and

Ahmad, 29).

59 This is clearer in TFS1, Bodleian ms., fols. 192b-193b/Digby Coll. ms., fol. 161b, than in

TFS,479, 480; for the Qinnawj campaign, see IB, III, 144 (tr. Gibb, 617).

60 TFS, 473; a clearer chronological indication at 482. I cannot agree with Nizami (in HN, 524),

who sees high taxation in the Doab as a response to famine in Delhi rather than as its ultimate cause.

61 TFS, 482. IB, III, 316 (tr. Gibb, 708): the remark, made in the context of the transfer of capital,

seems to apply to his initial entry into Delhi, rather than to a subsequent visit, which is why Husain, Rise

and fall, 121-3, and Tughluq dynasty, 171-3, dismissed it as based on hearsay.

~ships of the eastern Panjab, where the sultan was obliged to campaign against refractory peasants

later, in c. 738/1337-8.62

The sultan's efforts to encourage cultivation, after his return from the south, by having wells dug

in the vicinity of Delhi and by advancing seed and loans (sondhar) to peasants were unavailing.63

Campaigns into Katehr to plunder the grain for the use of his troops and of the people of Delhi were merely

short-term palliatives.64 Two years after his return to Delhi from the south, Muhammad was obliged to

permit a large-scale emigration from the capital to the fertile Awadh region, and himself set up a temporary

residence on the Ganges, at a locality named Sargadwari.65 His stay here of some two and a half years

seems to have alleviated the problems to some extent; and if Sirhindi is correct in claiming that the drought

lasted for seven years,66 the sultan's return to Delhi would have coincided with its end, i.e. c. 741/1340-1.

Measures to restore cultivation were still deemed necessary during the last years of the reign, although the

enormous cash advances to potential cultivators were not put to proper use and Barani believed that had

Muhammad returned alive from Sind the guilty parties would have been executed.67

We know that Muhammad's devaluation of the currency gave rise to a considerable degree of

inflation, entailing something like a fivefold rise in prices.68 The Sultanate's economic problems were

doubtless accentuated by the policy of the Chaghadayid khanate, since following his conversion

Tarmashirin had abolished those commercial duties not sanctioned by the Shari'a (mukus) and thus attracted

to Transoxiana merchants from Egypt and Syria in great numbers.69 This may have diverted a certain

proportion of the Egyptian trade north of the Hindu Kush, and might explain Muhammad's abolition of the

mukus within his own dominions (below, p. 272). The incentive could equally have been a general decline

in foreign trade as a result of the debasement; but it is significant that the Chaghadayid dinar enjoyed a

high reputation on account of its fineness.70 Possibly Muhammad's monetary policy had affected the balance of trade between India and Central Asia.

62 Malwa: TFS, 481-2. E. Panjab: TFS1, Bodleian ms., fol. 194b/Digby Coll. ms., fol. 162b; TFS,

483-4, refers only to the peasants' refusal to pay the kharaj, but does not link it with famine; IB, III, 372-3

(tr. Gibb, 734), for famine at Agroha.

63 TFS, 482, 484. IB, III, 299 (tr. Gibb, 700).

64 TFS\ Bodleian ms., fol. 195a/Digby Coll. ms., fol. 163a; TFS, 484-5, speaks merely of pasturage

(charakhur).

65 Ibid, 485-6. 66 TMS, 113. 67 TFS, 498-9. 'Afif, 92.1. Habib, 'Agrarian economy', 65-6.

68 Digby, War-horse, 39-40.

69 MA, ed. Lech, Ar. text 41 (German tr. 119). Al-Safadi, W'afi, X, 383, alone mentions the

abolition. For mukus, in origin customs duties, see generally W. Bjorkman, 'Males', End-Isl.' ■ P. G.

Forand, 'Notes on cusr and maks', Arabica 13 (1966), 137-41.

70 MA, ed. Lech, Ar. text 47 (German tr. 123).

~Military weakness and endemic rebellion

From Barani's testimony, it appears that prolonged unrest in the Doab acted as a spur to the next

wave of revolts in more distant provinces from c. 1334 onwards, notably those in Ma'bar, Bengal and

Tilang.71 The revolt of Sayyid Jalal al-Din Hasan, apparently kotwal of Madura, who assumed the title of

Sultan Jalal al-Din Ahsan Shah, was probably the first and is believed to have occurred in 734/1333-4.

Muhammad's representatives were killed, and the troops supposedly garrisoning the province did nothing.72

This crisis was closely followed by the loss of Bengal. Fakhr al-DIn (also known as 'Fakhra') was the

former armour-bearer (silahdar) of the sultan's adopted brother Bahram Khan, and had already made an

unsuccessful bid to seize power at Sunarga'un on his master's death. The rising was checked by Qadr Khan,

Muhammad's representative at Lakh-nawti; but not long afterwards a prolonged struggle broke out for

control of the province. First Qadr Khan's troops mutinied, slew him and went over to the rebel Fakhr al-DIn, who established his residence at Sunarga'un. Then Fakhr al-Din's lieutenant at Lakhnawti was killed

by Qadr Khan's former 'arid, 'Ali Mubarak, at the head of loyalist troops. When the sultan proved unable to

comply with his request that a new governor be dispatched from Delhi, 'Ali Mubarak found himself obliged

to assume the royal title himself as Sultan 'Ala' al-Din 'A1i Shah in order to rally support against the hostile

activities of Fakhr al-Din. Both 'Ali Shah in the middle of the 1340s and Fakhr al-DIn's son and successor,

Ikhtiyar al-Din Ghazi Shah, in the early 1350s would be overthrown by a third candidate for the

sovereignty, a former retainer (chakir) of 'Ali Mubarak named Ilyas Hajji, who reigned as Sultan Shams al-

Din.73 Like 'Ali Shah, Ilyas seems to have recognized the authority of Delhi, since a farmdn of

Muhammad's successor Firuz Shah

71 The link is explicit in the first recension: TFS1,Bodleian ms., fol. 193/Digby Coll. ms., fols. 161b-162a.

72 Meagre details in TFS, 480, except that Muhammad is said to have been campaigning around

Qinnawj when the news arrived. FS, 469 (tr. 705). IB, III, 144 (tr. Gibb, 617). For the date, see S. A. Q.

Husaini, 'The chronology of the first two sultans of Madura', Proceedings of the Pakistan History

Conference, 5th session, Khairpur 1955 (Karachi, n.d. [1958]), 193-7, and 'The history of the Madura

Sultanate', JASP 2 (1957), 91-5, citing a coin of Ahsan Shah dated 734. J. Burton-Page, 'Djalal al-Din

Ahsan', Enc.Isl.2, is therefore 7J in need of updating.

73 TMS, 104-5, provides the fullest account, though with incorrect dates. TFS, 480, and FS, 472 (tr. 709), are laconic. IB, IV, 213-14 (tr. Gibb and Beckingham, 869), garbles the details and does not

mention Ilyas, on whom see SFS, 47 (tr. Basu, JBORS 27 [1941], 92); this last source calls 'Ali Shah the

armour-bearer of Dinar, one of Qadr Khan's eunuchs. For a survey of events, see Abdul Karim,

'Circumstances that led to the independence of Bengal (1338 A.D.)', Proceedings of the Pakistan History

Conference, 5th session, 209-22. Coins of All Shah go down to 744: CCIM, II, 150 (nos. 22-3). GhazI Shah

struck coins in 751: ibid., II, 149 (no. 21). Ilyas had begun to reign by 743/1342: Dani, 'Shamsuddin Ilyas

Shah', 55; Eaton, Rise of Islam, 86.

~Would later claim that he had remained submissive until after Muhammad's death.74

It may have been the presence of actively loyal troops in Bengal that induced Muhammad to give

priority to the suppression of Ahsan Shah in Ma'bar. At the head of a sizeable force, he moved south in 735/1334-5 and passed through the Deccan. But on its arrival in Tilang, the army was struck by some kind

of epidemic (wuba), and the sultan was obliged to retreat; he himself fell gravely sick when he reached

Dawlatabad, recovering only after his return to Delhi. That the campaign had been a major disaster was

apparent to Ibn Battuta, who dates from this juncture the falling-away of outlying provinces.75 The failure

to recover Ma'bar gave the signal to other would-be dissidents, and encouraging rumours of Muhammad's

death circulated widely. Already, as the sultan marched southwards, one of his officers, Taj al-Din Hushang

(the son of Kamal al-Din 'Gurg'), muqta' of Hansi, fled to the Vindhyas and thence into the Konkan;

Qutlugh Khan, Muhammad's old tutor and governor of the Deccan, moved against him and eventually

induced him to yield with a promise of safe-conduct.76 Around the same time a Mongol commander named

Hiilechu occupied Lahore in alliance with the Khokhar chief Gul Chand, the one-time ally of Muhammad's

father; the rebels were defeated and the city retaken by the wazir Khwaja Jahan.77 The seizure of Multan by the Afghan chief Shahu, which 'Isami makes part of this insurrection in the western Panjab, is treated by

other sources as a separate episode. Muhammad, who had now returned to Delhi, viewed this revolt as

sufficiently threatening to warrant dealing with it himself; but Shahu made off on his approach and sent a

message of submission.78 More serious were the loss of Kampila, which now became the nucleus of the

kingdom of Vijayanagara, and a rising in Tilang, whence the governor, Malik Maqbul, was expelled by

Kapaya Nayak and fled to Delhi, arriving a matter of days after the sultan himself.79

The loss of Tilang, the province whose reduction during the previous reign had been his personal

achievement, dealt Muhammad an especially severe blow. He is said to have wanted to mount an

expedition to recover it, but to have been prevented from doing so because of the famine.80 If 'Isami is to be

trusted, half the army commanders and a third of the troops had perished in the epidemic;81 while the view

both of Barani and of Ibn 74 IM, 16.

75 IB, III, 334-5 (tr. Gibb, 717).

76 FS, 469-70 (tr. 706-7). TMS, 106. IB, III, 335-6 (tr. Gibb, 717-18).

77 FS, 471 (tr. 707-8), erroneously making this part of the same episode as Shahu's revolt (below).

IB, III, 331-3 (tr. Gibb, 716-17).

78 TFS, 482-3. IB, III, 362 (tr. Gibb, 729). 79 Brief reference to both revolts in FS, 606 (tr. 902). Tilang: TFS, 484. For the limited material on

the emergence of Vijayanagara, see Husain, Tughluq dynasty, 248-9.

8o TFS1 Bodleian ms., fol. 195a/Digby Coll. ms., fol. 163a, says that he was inwardly (dar

batin) afflicted. TFS, 484, is briefer. 81 FS, 469, 471 (tr. 706, 708). IB, III, 334 (tr. Gibb, 717), says

merely that the greater part of

~Battuta's informants was that the Qarachil campaign had gravely weakened the army of the Sultanate.82 We have here the two circumstances that bedevilled Muhammad's government for several

years to come: a heavy reduction in the number of troops at his disposal, combined with a considerable loss

of revenue owing to a decline in cultivation, so that the sultan was unable to rebuild his forces.

The revolts of the middle period of the reign that we have considered so far smack of opportunistic

responses to a prolonged crisis, whether on the part of disaffected amirs or by Hindu elements on the

periphery of the Sultanate. But are there any signs of a deeper malaise affecting the ruling class itself? In

contrast with the system that obtained in the Mamluk empire, there was now a direct link between the

imperial treasury and the ordinary trooper, and the amirs had lost the capacity to bind troops to their own

interests with iqta' grants from their assignments, which were intended exclusively for their personal

maintenance.83 In addition, Ibn Battuta reveals that the military command had become completely

separated from the fiscal administration of the iqta' so that within the territory of Amroha, for instance, a wali al-kharaj, responsible directly to the sultan, is found alongside the amir.84 This assault on the position

of provincial commanders, it has been plausibly suggested, was one factor underlying the revolts in Gujarat

and the Deccan that plagued the sultan's last years.85

Loss of revenue accompanying the secession of a number of major provinces also had the

insidious effect of increasing pressure on Muhammad to demand larger sums from the regions that

remained loyal. Officers who had entered into contracts for the farming of revenue seem to have

undertaken to transmit unrealistically high sums to the sultan. Ibn Battuta was told of a Hindu who

contracted to farm the revenues of the entire Deccan province for seventeen krors (170,000,000 tangas),

but was unable to meet his obligations and was flayed alive on Muhammad's orders.86 The story cannot be

tied in with any episode recounted elsewhere, but it illustrates the impact that such arrangements made on contemporaries. The impossibility of supplying the government's needs in this fashion could at times

engender rebellion by hitherto loyal servitors. Two risings which occurred during Muhammad's stay at

Sargadwari fall into this category. Nizam Ma'in, who farmed the revenues of Kara, and Shihab Sultani,

styled Nusrat Khan, who had undertaken to extract one kror (10,000,000 tangas) from Bidar and its iqta's

over three years, were both

the army that had accompanied the sultan perished. Cf. also FS, 472 (tr. 709), where the loss of

provinces is put down to Muhammad's lack of troops, and 510-11 (tr. 759).

82 IB, III, 327 (tr. Gibb, 714). TFS, All, 478.

83 MA, ed. Spies, 13 (German tr. 37, 38)/ed. Fariq, 24, 25 (tr. Siddiqi and Ahmad, 37-8). For

conditions in Ayyubid and Mamluk Egypt, see Rabie, Financial system, 32-8.

84 IB, III, 436, 439 (tr. Gibb, 762, 763). Conermann, Beschreibung Indiens, 146, 147-8.

85 Habib, 'Agrarian economy', 71-3.

86 IB, IV, 49 (tr. Gibb and Beckingham, 795).

~pushed into rebellion by their failure to raise the sums promised; Nusrat Khan is said to have

been unable to recover even a third or a quarter of the farm. Nizam Ma'in's feeble bid for independence was

snuffed out by the sultan's governor of Awadh, 'Ayn al-Mulk Ibn Mahru, and his brothers. Nusrat Khan was dealt with by the ubiquitous Qutlugh Khan, who gathered troops from Dawlatabad but eventually

persuaded him to surrender under guarantee of safe-conduct.87

There are other signs that Muhammad's regime was becoming the prisoner of its own reputation

for harshness. Barani asserts more than once that the uncompromising punishments inflicted at Delhi

occasioned fear and disaffection elsewhere in the empire, which played their own part in fomenting

revolt.88 The rising of 'Ayn al-Mulk Ibn Mahru in Awadh provides an illustration. Suspecting that Qutlugh

Khan's officials were embezzling some of the tax revenues in the Deccan, the sultan contemplated recalling

his old tutor and transferring to Dawlatabad Ibn Mahru, who had recently demonstrated his loyalty and

efficiency by shipping large quantities of grain and other goods from Awadh to Sargadwari and Delhi at the height of the famine. In his first recension, Barani has Muhammad eagerly anticipating the increased sums

that an administrator of Ibn Mahru's calibre might obtain from the much wealthier Deccan.89 Unfortunately,

the sultan also learned that large numbers of Delhi's residents had fled from the capital to Awadh, attracted

by its prosperity and by Ibn Mahru's mild government, and demanded that they be sent back. Ibn Mahru,

who was warned of Muhammad's anger over this, inferred that the planned transfer to the Deccan was

simply a ruse to dispose of him, and he and his brothers decided to pre-empt their execution by rebellion.

Muhammad defeated them on the Ganges, not far from Qinnawj. Ibn Mahru's brothers were killed in the

fighting or disappeared, and he himself was taken prisoner; but it is a measure of his stature, and of the

sultan's understanding of the reasons for his revolt, that he was not long afterwards restored to favour.90 He

was later appointed governor of Multan at the time of Muhammad's final campaign against the rebel Taghai

and his Sumra allies in Sind.91

87 TFS,487,488.

88 Ibid., 472, 484, 499-500, 517.

89 TFS1, Bodleian ms., fol. 196a/Digby Coll. ms., fol. 163b.

90 IB, III, 341-54, 357 (tr. Gibb, 720-6, 727), provides a detailed account of the campaign, in

which he participated, but is unaware of the impulses behind the revolt. So too is FS, 472-5 (tr. 709-14).

TFS1 Bodleian ms., fols. 195b-196a/Digby Coll. ms., fol. 163, and TFS, 486-7, 489-91, analyse Ibn

Mahru's motives. The suspicions about revenue from the Deccan are mentioned only in TFS, 500-1: Hardy,

'Didactic historical writing', 53-5, compares the two recensions at this point. There is a brief account of the revolt in TMS, 109-10. 'Afif, 406-8, for Ibn Mahru's restoration to favour.

91 IM, 106, 107. Despite the doubts expressed by Abdur Rashid in his introduction (27), Ibn

Mahru's reference to the sultan's having spared him makes it certain that these two letters date from

Muhammad's reign rather than that of Firuz Shah. He also says that at the time of his appointment he has

been ordered to supply troops and ships, which places the date of the letters around the time that

Muhammad crossed the Indus not long before his death:

~describing how Muhammad's forces took up position near Qinnawj for the encounter with Ibn

Mahru, Ibn Battuta alludes to the antipathy between the rebel, who was of Indian extraction, and the 'amirs

of Khurasan and foreigners' accompanying the sultan.92 From c. 734/1333-4, as we have seen, Muhammad

was intent on making clients of local rulers in Khurasan and neighbouring regions and thus achieving by means of patronage what he had been unable to accomplish through the Khurasan project. This was in turn

part of a wider policy of favouring foreigners over the indigenous aristocracy (above, pp. 184-5). Whether

this in itself was enough to incite members of the Indian Muslim aristocracy to revolt, we cannot know; but

it may well have played a role in the unrest of the sultan's later years. We are perhaps on surer ground in

identifying resentment towards Muhammad's pagan Hindu agents as one of the mainsprings of disaffection.

In c. 1341 there was a rising in Siwistan, in which the local Sumra ruler Unar (Ibn Battuta's 'Wunar') and a

military officer named Qaysar-i Rumi slew the Hindu bureaucrat 'Ratan' whom the sultan had appointed as

muqta' of the province. Unar soon deserted his associates, and Qaysar and his followers were put down

without difficulty by the governor of Multan, 'Imad al-Mulk Sartiz.93 Similarly, as we shall see, Bhiran, the

pagan muqta' of Gulbarga, would be the first victim of the rising of 'A1i Shah Kar in the Deccan.94 Yet

until the middle of the 1340s Muhammad seems to have retained the support of the military class as a whole. Barani describes Nusrat Khan as a grain-dealer (baqqal), and scornfully contrasts Ibn Mahru and his

adherents, who were 'clerks and grain-dealers' (nawisan-dagan-u baqqalan), with the sultan and the

seasoned troops who had served both him and his father Tughluq and who could hardly have been expected

to desert him.95

The implication is that Muhammad was a military man's sovereign.

Muhammad took various steps during the early 1340s that were undoubtedly designed to rally

support behind his regime. Although Barani gives the impression that the sultan entered into relations with

the puppet 'Abbasid Caliphate at Cairo during his stay at Sargadwari,96 we know from Egyptian sources

that he had already been in contact with the Caliph al-Mustakfi bi'llah as early as 731/1330-1, and that at

least three embassies had been

TFS, 523. The date 9 Shawwal, when Ibn Mahru was despatched to Multan, must therefore belong

to 751 (10 December 1350). Nevertheless 'Afif makes out that Firuz Shah appointed Ibn Mahru to the

province (below, p. 303).

92 IB, III, 344, 349 (tr. Gibb, 721-2, 724).

93 Described only ibid., Ill, 105-8 (tr. Gibb, 599-600). This episode must have fallen not just prior

to IB's arrival in India, as its place in the narrative suggests, but before the visit to Slwistan in 742/1341

mentioned later, III, 447 (tr. Gibb, 766-7). One reason for dating the Slwistan revolt this late is that Sartiz

was not appointed to Multan until after Shahu's rising, i.e. c. 1337 (above, p. 183, n.86).

94 TFS, 488. Cf. also FS, 485-6, 487-8 (tr. 726-8, 730-1). Nizami (in HN, 565) reaches similar

conclusions about the role of hostility towards the sultan's Hindu servitors.

95 TFS, 488, 490. 96 Ibid., 491-2.

~sent from Delhi.97 In 741/1340-1 Muhammad substituted the name of al-Mustakfi (who had in

fact died in the previous year) for his own on the coinage and in the khutba.98 It was in this same year,

according to Ibn Battuta, that he abolished all uncanonical taxation (mukus);99 and perhaps also that he took

to presiding in person over the mazalim tribunal for the redress of his subjects' grievances.100

Not until 746/1345-6 did Muhammad's envoy, Hajji Rajab Burqu'i, return to Delhi with the personal robe of the Caliph al-Hakim bi-amri'llah, al-Mustakfi's son and successor, and a diploma

conferring on the sultan the rank of the caliph's lieutenant; he was accompanied by the Egyptian grand qadi,

the shaykh al-shuyukh Rukn al-Din al-Malati, head of the convent of Siryaqus. In the meantime, in

744/1343, an unofficial envoy from Cairo, Hajji Sa'id Sarsari, had brought Muhammad a diploma, a banner

and a robe. The ceremonial surrounding these occasions, when the sultan adopted a stance of extreme

humility, clearly made a powerful impression on Barani.101 By such propagandistc gestures the sultan

hoped, perhaps, to recover the support of the 'ulama' and others of the 'religious class' and hence,

presumably, to legitimize his position in the face of would-be rebellious amirs.

Confrontation with the amiran-i sada

Within the next year or two, however, the situation once again deteriorated. Muhammad had come to believe that the local commanders in Gujarat and the Deccan, the amirs of a hundred (amiran-i sada),

were responsible for the fiscal problems of his government, and decided to supersede them by bringing the

revenues of the two provinces under closer control by the centre. According to Barani, the sultan in

745/1345 believed that large sums

97 Jackson, 'The Mongols and the Delhi Sultanate', 131-2 n.74.

98 CMSD, 122-4 (nos. 491-491H), 147-8 (nos. 617-622A).

99 IB, III, 288 (tr. Gibb, 694), for the abolition of mukus; at III, 117 (tr. Gibb, 605), this is said to

have coincided with Muhammad's recognition of the caliph, but is dated two years after IB's arrival in India, i.e. c. 736/1335-6. A list of taxes abolished in Muhammad's reign -manduh, *tarka, mal-i mawjud,

chahar bazar, dard'ib, gudharha and khardj-i muhtarifa-yi muslim - is given in IM, 79; for those abolished

by Firuz Shah, see SFS, 124; FFS, 5 (tr. Roy, 453); I. H. Qureshi, The administration of the Sultanate of

Dehli, 4th edn (Karachi, 1958), 244-7 (appendix H).

100 IB, III, 288-9 (tr. Gibb, 694-5).

101 TFS, 492-6; and cf. also 460. There is a fuller and clearer account in SFS, 280-2; tr. in Shaikh

Abdur Rashid, 'Firuz Shah's investiture by the Caliph', MIQ 1 (1950), 69. See also IB, I, 363-70, and III,

248-9 (tr. Gibb, 225-8, 674), who distinguishes the status of the two envoys Hajji Sa'id and Hajji Rajab. For the arrival of Rajab's party in Cairo in 744 and for Rukn al-Din, see al-Shuja'i, Ta'rikh al-Ndsir Muhammad

ibn Qala'uni'l-Salihi wa-Awladihi, ed. and tr. Barbara Schafer, Die Chronik as-Sugd'i's, QGIA, II

(Wiesbaden, 1977-85, 2 vols.), I (text), 257-8, and II (tr.), 290-1; the date Rabi I 743 for Rukn al-Din's

departure from Cairo given by al-Maqrizi (d. 845/1442), al-Suluk li-Ma'rifat Duwali' l-Muluk, ed. M.M.

Ziada et al. (Cairo, 1934- in progress), II, part 3, 887, must be an error.

~had been held back for years by the officials in Bharuch;102 and in his account of the revolt of

'Ali Shah Kar ('the Deaf ') at Bidar, 'Isami, who is particularly well informed about the Deccan and

accordingly furnishes much greater detail than other writers, suggests that the rebellion was sparked off by

new revenue-raising arrangements.

'A1I Shah, a Khalaj officer and a nephew of Sultan 'Ala' al-Din's general Zafar Khan, is described

as an amir-i sada of Qutlugh Khan103 who had rendered signal service by fighting against Nusrat Khan and by reducing the district of Koyir. He continued to serve faithfully and transmitted the stipulated monies,

until a Hindu named Bhiran, who held the iqta' of Gulbarga, grew aware of the sums being retained from

Koyir and prevailed upon Qutlugh Khan to let him farm the revenues, undertaking to increase them by 50

per cent. 'A1i Shah reacted by seizing Bidar and Gulbarga and killing Bhiran, and assumed the royal title as

Sultan 'Ala' al-Din. After some time Qutlugh Khan, aided by reinforcements from the sultan, was able to

induce him to surrender.104 His uprising, which occurred in a region whose officers Muhammad viewed

with suspicion, looks like a localized rehearsal for the wider insurrection against the sultan during his last

years.

At some point early in 745/in the spring-summer of 1344 the sultan took the decision to separate

the enormous Deccan province currently supervised by Qutlugh Khan into four divisions (shiqqs). Qutlugh Khan was to be recalled and replaced as wazir at Dawlatabad by Imad al-Mulk Sartiz, hitherto governor of

Multan; in the interval the command at Dawlatabad was to be exercised by Qutlugh Khan's brother 'Alim

al-Mulk Nizam al-Din, the governor of Bharuch. According to Barani, the men chosen to command the

shiqqs all had a reputation for shedding blood, and both he and 'Isami allege that the people of the Deccan,

who had come to regard Qutlugh Khan's regime as a safeguard against the ordeals experienced in

Muhammad's other territories, were dismayed at the amir's departure.105

The principal target of the new administration, however, was the amiran-i sada. Barani says that

the men sent from Delhi were under instructions from the sultan to regard these officers as the chief

instigators of unrest. We might be tempted to discount reports of Sartiz's previous exactions in the

102 TFS, 513. 103 Thus only ibid., 488, where he is described as the son of Zafar Khan's sister (as also in TMS,

108); though TFS, 508, calls him the son of a brother.

104 By far the most detailed account in FS, 483-500 (tr. 725-47); see 479 (tr. 718-20) for his

service against Nusrat Khan. The information in TFS, 488-9, and IB, III, 357-8 (tr. Gibb, 727-8), is limited.

Barani dates the revolt during Muhammad's stay at Sargadwari, whereas IB places it after his return to

Delhi.

105 TFS, 501-2. FS, 503 (tr. 749-50); and see also 462 (tr. 696-7), where the security of the people

of the Deccan, however, is attributed ultimately to the presence there of the saint Shaykh Zayn al-DIn. IB, III, 336-7 (tr. Gibb, 718), comments on the confidence inspired by Qutlugh Khan and on his liberality. The

date of the order recalling Qutlugh Khan to Delhi is given as 1 Sha'ban 745/8 Dec. 1344 by Badr-i Chach,

Qasa'id, ed. Hadi 'Ali, 64/ lithograph ed. M. 'Uthman Khan (Rampur, 1872-3, 2 vols.), II, 407.

~Multan province, which according to the correspondence of his successor there, Ibn Mahru, had

still not recovered a few years later.106 But it is surely no accident that 'Aziz Khammar, the sultan's newly

appointed governor of Dhar and one of the four shiqq-commanders, had made a name for himself as an

oppressive revenue-collector in the Amroha district, in which capacity he had clashed with the local

military commander.107

Now, soon after his arrival at Dhar, 'Aziz summarily executed some eighty amiran-

i sada. When this news reached their counterparts in Gujarat and the Deccan, they rose in revolt.108 Ibn Battuta likewise mentions instructions for the killing of military commanders, but he makes the revolt start

in Gujarat, where Malik Muqbil allegedly received orders to put them to death. We should not necessarily

accept the testimony of Ibn Battuta (who does not employ the term 'amirs of a hundred') that the victims

were all Afghans: he seems to have been misled by the fact that the leaders of the ensuing revolt - QadI

Jalal in Gujarat and Isma'il *Mukh in the Deccan - both belonged to that race. He is certainly in error,

moreover, in linking Muhammad's orders to massacre 'Afghans' with his campaign against the Afghan

Shahu in Sind, which had occurred some eight years or so prior to these developments (see above, p.

268).109

The atrocity perpetrated by 'Aziz Khammar turned the explosive situation in the south into one of

open rebellion. Whereas hitherto Muhammad had been confronted by the recalcitrance of individual

grandees and their retinues, he now faced a widespread insurrection embracing the officer class in two major provinces. When the news reached the amiran-i sada in Dabhoi and Baroda, they attacked and

routed Malik Muqbil, the na'ib-wazir of Gujarat, and plundered a convoy of treasure he was escorting on its

way to Delhi. Kanbhaya was surrendered to them, and they were able to take Asawul.110 'Aziz Khammar,

who moved against them from Dhar and was joined by Muqbil, was defeated and captured by the

insurgents and put to death. Returning to Kanbhaya, Qadi Jalal and his adherents settled

106 IM, 78-9, 88.

107 IB, III, 436-40 (tr. Gibb, 762-3): this seems to have transpired at the time of Muhammad's

absence on the Ma'bar campaign.

108 TFS1, Bodleian ms., fols. 202b-203a/Digby Coll. ms., fol. 168a, is more explicit here than

TFS, 503-4, 507. IB, III, 364 (tr. Gibb, 730-1), also links the revolts in Gujarat and the Deccan. For these

revolts, see generally I. Prasad, Qaraunah Turks, 208-53; Husain, Tughluq dynasty, 283-97; HN, 540-55.

109 FS, 504 (tr. 750), for royal orders. IB, III, 362, 364-6 (tr. Gibb, 729, 730-1). TFS, 514, at one

point appears to support the equation, by speaking of the rebels at Dawlatabad as 'these Afghans'; but BL

ms., fol. 254a, reads not in afghAnan but in afghan, i.e. the singular, denoting the leader Isma'il *Mukh.

SFS, 20, calls the rebel officers in Gujarat simply 'army chiefs' (saran-i guruh). For other Afghan officers

in the Gujarat and Deccan revolts, see Siddiqui, 'The Afghans and their emergence', 255-6.

110 FS, 503-6 (tr. 750-3), provides the fullest account; and see also TFS, 507. TMS, 111, gives a

brief notice (sub anno 748 in error). IB does not mention the plundering of the convoy.

~down to besiege the city.111 Muhammad, who had been preparing to head an army against the

rebels since learning of the attack on Muqbil in the latter half of Ramadan 745/late January 1345, halted at

Bharuch. Here he instituted oppressive measures for the extraction of the arrears of revenue, ordering Malik

Maqbul, the na'ib-wazir of the empire, who had pursued the enemy as far as the banks of the Narbada, to

kill the amiran-i sada of Bharuch under his command. The back of the Gujarat revolt appeared to be

completely broken. Qadi Jalal and his lieutenants narrowly escaped being handed over to the sultan by

Nanadeva ('Man Deo'), the Hindu raja of Baglana, Salher and Mulher, and fled to Dawlatabad, where the

amirdn-i sada were by now similarly in arms against the sultan.112

In the Deccan Muhammad's policies had provoked a major crisis. Two of the sultan's principal agents were known to be already on their way to Dawlatabad to conduct an inquiry into the loyalties of the

province; and in addition the sultan sent two other amirs to 'Alim al-Mulk Nizam al-Din with instructions

to have the more important amiran-i sada of that region brought under guard to Bharuch. The proposed

victims of the purge set off from Dawlatabad, but realized Muhammad's intentions and turned back. 'Alim

al-Mulk was arrested, and the two royal agents were put to death. The province fell under the control of the

amirdn-i sada; Ismail *Mukh, a brother of the Afghan Malik Malik who had held a command in Tughluq's

reign, was proclaimed sultan as Nasir al-Din.113

On receipt of this news, Muhammad advanced by forced marches to the Deccan and inflicted a

heavy defeat upon the rebels. He took up his quarters in the royal palace at Dawlatabad, and his troops invested the fortress of Dharagir, where Ismail *Mukh and his chief adherents had taken refuge.114 But as

the sultan busied himself with setting the affairs of the region in order, he received reports of a fresh rising

in Gujarat, led by a Turkish slave named Taghai. Taghai, formerly Muhammad's shihna-yi bargah, had

been banished by the sultan to the Yemen as a punishment for some misdemeanour, but had been caught up

in the fighting at Kanbhaya while awaiting embarkation. Having played a crucial role in the city's defence

against Qadi Jalal, he was restored to favour. In the sultan's absence, however, he fell out with

Muhammad's lieutenant at Asawul, Tatar Malik, and made common cause with the amirdn-i sada of

Gujarat. The rebels entered Nahrwala, slew its governor, sacked Kanbhaya, and laid siege to Bharuch.

When

111 FS, 506-10 (tr. 753-9). TFS, 509, reports the sultan's receipt of the news of 'Aziz's defeat and

death. IB, III, 364 (tr. Gibb, 730), is very brief. 112 FS, 512-14, 522 (tr. 760-4, 773-4). TFS, 511-13, has the rebels being defeated by Malik

Maqbul near Dabhol and Baroda; Barani alone describes Muhammad's conduct at Bharuch. For the identity

of 'Man Deo', see Hodivala, Studies, I, 299.

113 TFS, 512, 513-14. A fuller account in FS, 516-21 (tr. 766-73). IB, III, 365-6 (tr. Gibb, 731), is

relatively brief and calls the rebel leader Malik Mall's son in error.

114 FS, 530-6 (tr. 783-91). TFS, 514-15. IB, III, 368-9 (tr. Gibb, 732-3): this is his latest

information on the revolt.

~

Muhammad advanced on Bharuch, Taghai fled to Kanbhaya, where he defeated a force that the

sultan had sent in pursuit and killed its commander, Ytisuf-i Bughra, before taking flight again when

Muhammad hurried after him.115

While receiving the submission of various local ranas and chiefs in Girnar (Junagadh; the modern

Kathiawad), Muhammad was recalled to the Deccan by the news that his amir 'Imad al-Mulk Sartiz, whom

he had deputed to reduce Gulbarga following the victory over Ismail *Mukh, had been defeated and killed

by another group of amiran-i sada under one of Ismail's lieutenants, Hasan Gangu, styled Zafar Khan. The

troops the sultan had left at Dharagrr had fallen back on Dhar, and Hasan Gangu had made a triumphal

entry into Dawlatabad. Isma'il *Mukh renounced the royal title in favour of his deliverer: Hasan Gangu, who was enthroned on 24 Rabi' II 748/3 August 1347 as 'Ala' al-Din Bahman Shah, thereby became the

first sovereign of the independent Bahmanid dynasty that ruled in the Deccan until the sixteenth century.

According to Barani, Muhammad summoned various commanders from Delhi and planned to send them to

regain the Deccan, but abandoned the idea when he heard reports of the great numbers rallying to Hasan

Gangu's standard. It seemed advisable to deal first with Taghai and to postpone turning his attention to the

south until later.116

Muhammad spent the next three monsoons in ineffectual pursuit of Taghai, moving from

Nahrwala to Kathiawad and back again, before making an abortive attempt to assault Thatta, with whose

Sumra princes the rebel had taken shelter. The sultan was preparing for a second attack on Thatta when he

fell ill and died on the banks of the Indus on 21 Muharram 752/20 March 1351.117 His achievements during these last years should not be underestimated. By concentrating on the overthrow of Taghai, a task which

was not in fact completed in his lifetime, he had at least accomplished the subjugation of Gujarat -

including regions that do not seem to have acknowledged his predecessors - and ensured that the province

remained part of the Sultanate for another two generations. But any larger enterprise was beyond the

depleted resources at his disposal. An alleged conversation between the sultan and Barani, in which

Muhammad complained that a new revolt erupted in one direction every time he turned towards another,

115 There is valuable information on the origins of Taghai's revolt in SFS, 19-21, 23-4 (tr. Basu,

JBORS 23 [1937], 97-102). TFS, 515-16, 517-20, gives a narrative. FS, 538-9 (tr. 793-4), is brief.

116 TFS 52O-I, 522, but giving only a short notice of events in the Deccan; see also 515 for the

despatch of Sartiz towards Gulbarga. FS, 540-54 (tr. 811-28), provides a full account of Hasan Gangu's

operations, with the date of his accession.

117 His movements have now been elucidated, on the basis of material in TFS1, by Simon Digby,

'Muhammad bin Tughluq's last years in Kathiavad and his invasions of Thattha', in Khuhro (ed.), Sind

through the centuries, 130-8.

~conveys his exasperation.118 The disaffection of the amiran-i sada, which Muhammad himself

had done so much to foster, was too intense and geographically too widespread to be overcome, given the

sultan's fiscal problems and the decline in the impressive military establishment he had presided over in the

early years of his reign. 118 TFS, 521. FS, 538 (tr. 794), similarly catches the dilemma confronting Muhammad at

Dawlatabad when he first heard news of Taghai's rebellion.

~CHAPTER 14

The sultans and their Hindu subjects

The Delhi Sultans were first and foremost Islamic rulers. Fakhr-i Mudabbir calls Iltutmish 'the

sovereign of Islam' (padishah-i Islam).1 Juzjani saw Nasir al-Din Mahmud Shah as the 'Sultan' (or 'Sultan

of the Sultans') 'of Islam', or as 'Emperor of the Peoples of Islam'.2 Alternatively, the monarch could be hailed as 'Sultan of the Turks and Persians ('Ajam)'3 - ruler, in other words, over the war-lords, soldiers and

scholars who made up the immigrant Muslim population. In the eyes of the Sultanate's chroniclers, the

Muslims constituted what in more recent times would be termed a Staatsvolk. The monarch was

emphatically not sultan of the Hindus or of, say, the people of Hariyana; it has been observed that in our

Muslim sources Hindus 'are never interesting in themselves, but only as converts, as capitation tax-payers,

or as corpses'.4 All the sultans with one exception proclaimed the spirit in which they approached their task

by assuming on their coins and in their inscriptions the style (kunya) of Abu'l-Muzaffar ('Father of the

Victorious One'); the exception, Muhammad b. Tughluq, styled himself al-Mujahid fi Sabili'llah ('The

Warrior in the Path of God'). For many Muslim observers, the ultimate justification for any ruler within the

Islamic world was the protection and advancement of the faith. For the sultans, as for their Ghaznawid and

Ghurid predecessors, this entailed the suppression of heterodox Muslims, and Firuz Shah attached some

importance to the fact 'that he had acted against the ashab-i Ilhad-u Ibahat ('deviators and latitudinarians').5 It also involved plundering, and extorting

1 AH, 15.

2 TN, I, 273, and II, 91, 166, 185 and n.3; cf. also II, 91, padishah-i ahl-i ayman, 205, padishah-i

Musulmanan.

3 Taj, fol. 217b. TN, I, 297 (tr. 231); cf. also I, 275, 366 (tr. 183, 388). Shu'aib, 'Inscriptions from

Palwal', 2, 3; Yazdani, 'Inscriptions of the Turk Sultans', 15; RCEA, X, 72-3 (no. 3703).

4 Hardy, Historians, 114. 5 FFS, 6-8 (tr. Roy, 454-6). SFS, 129 ff. See also KF, 20, and TFS, 336, for 'Ala' al-Din's treatment

of ashab-i Ibahat. Qaramita - probably Isma'ilis - had attempted a coup in Delhi during Radiyya's reign:

TN, I, 461-2 (tr. 646-7); FS, 122 (tr. 236-7), seems to date this to Iltutmish's era.

~tribute from, independent Hindu principalities. That the Muslim ruler had a further duty to

eradicate infidelity and humiliate his Hindu subjects was a view expressed with particular frequency and

stridency by Barani, both in the Ta'rikh-i Firuz-Shahi and in his Fatawa-yi Jahandari, a mirror for princes

composed a few years later.6 To what extent these policies were actually implemented within the Sultanate

is the question to which we now turn.

Hindus in the service of Islam

It comes as no surprise to find Hindus carrying on their normal avocations in the service, and for

the benefit, of their Muslim rulers. The Turko-Persian nobility in the thirteenth-century Sultanate

accumulated enormous debts to Hindu bankers and brokers, the 'Multanis' and sdhs, who could still be

numbered among the sultan's wealthiest and most important subjects in the wake of 'Ala' al-Din Khalji's

economic reforms.7 A Hindu chieftain, Sadharana, is said to have served as 'Ala' al-Din's treasurer.8

Moving down the social scale, the sultans depended, for their ambitious construction projects, on a host of

Hindu labourers (70,000 of them in the service of 'Ala' al-Din, if we can believe Barani),9 who were

doubtless usually slaves. But these projects also relied on the expertise of a lesser number of master

craftsmen, like 'Mokha Mehta, son of Keta Mehta the Indian' who is commemorated in an inscription dated 740/1340 in the mosque at Baroda,10 and the masons recruited to repair the Qutb Minar.11 Such skilled

artisans seem to have been rewarded with immunities, as was the Hindu carpenter to whom the sultan's

governor of Bijapur in 1320 granted an estate free of taxes and other incidents for his services in the

construction of the great mosque.12 Members of the Hindu clerical class, too, were needed to staff the

administration, even if under the supervision of Muslim ministers and officials: a Hindu clerk in the service

of Qutb al-Din Khalji wrote a treatise on the operation of the mint in 1318,13 and another, Gujar Sah, was

responsible for overseeing the introduction of a new coin, the shashgani,

6 See especially FJ, 165.

7 TFS, 120, 284. For the Multanis, see further ibid., 311, 385; and for the sahs of Deogir, IB, IV,

49 (tr. Gibb and Beckingham, 794-5).

8 Pandit Ram Karna, 'Ladnu inscription of Sadharana of Vikrama samvat 1373', El 12 (1913-14),

19 (verse 13).

9 TFS,341.

10 ARIE (1963-4), 125 (no. 85); for the Sanskrit portion of this epigraph, see ARIE (1961-2), 143

(no. 1311).

11 P. Prasad, Sanskrit inscriptions, 21-2, 32-5 (nos. 1:9 and 1:14).

12 Nazim, Inscriptions of Bijapur, 25.

13 V. S. Agarwala, 'A unique treatise on medieval Indian coins', in H. K. Sherwani (ed.), Dr.

Ghulam Yazdani commemoration volume (Hyderabad, AP, 1966), 87-101. G. H. Khare, 'Dravyapariksha of

Thakkura Pheru - a study', JNSI28 (1966), 25-37.

~under the Tughluqid Firuz Shah.14 The sultans relied, lastly, on members of the Hindu menial

class for the execution of Muslim and Hindu rebels alike.15

It was the same in the military sphere, where the sultans, like their Ghaznawid predecessors,

maintained bodies of Hindu as well as Turkish troops. The slave infantry-guards and paiks in the sultan's entourage may well have come to enjoy the kind of privileged status that had belonged to Turkish ghulams

for much of the thirteenth century. Prior to the Khaljl era the evidence is sketchy: we know, for instance,

only that Balaban, prior to his accession, was attended by a body of a thousand paiks.16 'Ala' al-Din Khalji

is known to have recruited some two thousand paiks at Kara for his expedition to Deogir in 695/1296,17

and

they presumably remained in his service when he became sultan later that year. In the face of a bid by his

nephew Ikit Khan to kill him and seize the throne, 'Ala' al-Din was defended by an Indian slave, Nanak

(subsequently raised to the rank of malik), and by his paiks.18 When after his death his minister Malik

Kafur, himself a converted Hindu slave, set aside the late sultan's adult sons and tried to rule through one of

the younger princes, it was 'Ala' al-Din's old paiks again who in 715/1316 killed Kafur and secured the

throne for Qutb al-Din; although as a result, says Barani,.they started to give themselves intolerable airs and had to be suppressed.19 Qutb al-DIn himself, however, like his father, may have maintained a body of

paiks.20 Even the sternly Muslim Firuz Shah, whose mother was the daughter of a Bhatti chieftain from the

Panjab, employed members of his maternal kin: on one occasion, when his life was threatened by a

conspiracy, he was attended by his uncle, Rai Pheru ('Bhiru') Bhatti, who lent him his sword.21 In only a

few instances - that of Khusraw Khan, for example, in sharp contrast with his Parwari followers - do we

know that these Indian servitors converted to Islam.

Patronage of Hindus is associated particularly with the reign of Muhammad b. Tughluq. Jain

sources repeatedly mention the sultan's favour towards Jain scholars.22 Of the lowborn officers listed by

Barani whom Muhammad appointed to administrative positions, some were Hindus (above, pp. 185-6). It is

very probable that by Muhammad's death the position of the capital's Brahmans, at least, as representatives

of the Hindu population had been in some way regularized, since we are told that on Firuz Shah's entry into Delhi Brahmans were among those admitted to perform obeisance to the new sovereign;23 this is not

mentioned in connection with any earlier accession.

14 'Afif, 344-9.

15 E.g. TN, II, 82 (tr. 855); IB, III, 298, 339-40 (tr. Gibb, 700, 719-20).

16 TFS, 55.

17 Ibid., 222.

18 Ibid., 273. For Nanak, see above, p. 175.

19 TFS, 376-7.

20 Ibid., 392.

21 cAfif, 103-4.

22 Husain, Tughluq dynasty, chapter 11, with full references.

23 TFS, 546. 'Afif, 88.

~We have only meagre evidence for the attitudes of the Sultanate's Hindu subjects towards their

Muslim rulers. The significance of inscriptions in which the victorious (and sometimes fictitious) exploits

of the 'Saka' kings are extolled is open to question.24 An anecdote related by Ibn Battuta may carry greater

weight. He tells how a Hindu chief brought a charge against Muhammad b. Tughluq himself that he had

killed his (the chief's) brother without cause, and cited him to appear before the qadi. The sultan duly went,

unarmed and on foot, having in advance forbidden the qadi to show him any of the deference due to his

rank, and remained standing while the qadi gave judgement against him and ordered him to make

reparation to his accuser.25 This is an isolated instance, and the purpose of the story is to highlight the

sultan's humility and sense of equity; but it harmonizes with the general picture of Muhammad as a ruler

who, in the first half of his reign, took care to cultivate the Hindu. And if it embodies authentic fact, it

demonstrates that one Hindu, of some standing, recognized the authority of the Muslim qadi.26

As early as Iltutmish's reign, the sultans are soon found adopting practices that were distinctively

Indian, as for example riding elephants on ceremonial occasions, consulting astrologers and taking

horoscopes in advance of important occasions like an enthronement, and so on. Cultural borrowings of this

kind by Muslim rulers cannot be taken, of course, as a sign of accommodation with the infidel; they

represent merely an adaptation to Indian conditions (in much the same way as the first-generation

immigrants Fakhr-i Mudabbir and Juzjani employ the name of the Hindu month Ahar).27 Nevertheless, the

fact that in some degree they conducted their public lives in an Indian idiom may have facilitated the

acceptance of Muslim monarchs by Hindu chiefs.28

The problem of 'protection' and the jizya

Generally speaking, then, Hindus of diverse categories seem to have shown themselves

indispensable to the exercise of Islamic government and to the maintenance of Islamic institutions. But

what was their status under Islamic rule? According to the Shari'a, the 'people of the Book' (ahl al-kitab) -

those possessing scriptures which were seen as an inadequate expression of the truth contained in the

Qur'an - were to be treated as 'protected peoples' (ahl al-dhimma or dhimmis), once they had capitulated

and accepted Muslim 24 Palam baoli inscription of V.s. 1333/1276, in P. Prasad, Sanskrit inscriptions, 3-15 (no. 1:4).

25 IB, III, 285 (tr. Gibb, 692-3).

26 Hardy, 'Growth of authority', 194.

27 Horoscopes etc.: TN, I, 449 (tr. 623); TFS, 142, 456; FS, 393-4 (tr. 598-9); TMS, 79. Ahar: SA,

31 (text reads 'HA in error); TN, II, 21 (tr. 748).

28 Hardy, 'Growth of authority', 201, and 'Authority of Muslim kings', 49.

~government.29 The term 'people of the Book' was originally meant to apply to the monotheistic

Christians and Jews, but the mention in the Qur'an of a third, somewhat obscure people, the Sabians,

enabled the Muslim authorities to extend the category of dhimmis to the Zoroastrians in Iran. Dhimmis had

the right to practise their own faith, but they were not allowed to proselytize or to construct new places of worship. They were also subject to the jizy a, a capitation-tax in lieu of the military service performed by

adult male Muslims. In addition, at different times in different parts of the Islamic world rulers had

introduced discriminatory laws regulating the dress of dhimmis, forbidding them to ride horses or to bear

weapons, and so on. Muslim legal scholars differed over the rights dhimmis might enjoy, thus the Hanafi

school, which was dominant in the Sultanate, is alone in setting the blood-money for a dhimmi at the same

level as that for a Muslim.30

Whether the polytheists who confronted the Muslim conquerors within the Indian subcontinent

could be classed as a 'people of the Book' might appear at first sight to be a moot question. But in fact

Baladhuri's Futuh al-Buldan, one of the principal sources for the Muslim conquest of Sind in the early

eighth century, tells us that the Arab general Muhammad b. Qasim treated the idol-houses (budd) on a par

with Christian churches, Jewish synagogues, or Zoroastrian fire-temples.31 The term dhimmi was extended to embrace Hindu princes and their peoples who submitted and offered tribute, so that we read of the

acceptance of dhimma status by the inhabitants of Dvarasamudra in 711/1311-12, the ruler of Tilang in

718/ 1318, and the rai of Nagarkot in c. 766/1364-5.32 The list of those prepared to recognize the Sultanate's

Hindu subjects as dhimmis includes not merely Hasan-i Nizami, Juzjani, 'Afif, Ibn Mahru, 'Abd al-Hamid

Ghaznawi and the anonymous author of the Sirat, but also Barani, who as we shall see was by no means

well disposed towards even the submissive infidel, and the supposedly uncompromising Tughluqid Sultan

Firuz Shah in his Futuhat, drafted originally as an inscription and hence for public consumption.33 Even a

legal text of Firuz Shah's reign includes several references to dhimmis, by which it clearly means Hindus.34

29 For what follows, see Cl. Cahen, 'Dhimma', Enc Is!2; Bernard Lewis, The Jews of Islam

(Princeton, 1984), chapter 1 (for jizya, see especially 14-16). 30 E. Tyan, 'Diya', Enc.Isl2, III, 341. For a classic restatement of the disabilities to which dhimmis

were subject according to the 'Covenant of 'Umar', see Sayyid 'A1i Hamadani (d. 786/1385), Dhakirat al-

Muluk, ed. Sayyid Mahmud Anwari (Tabriz, 1358 Sh./1979), 285-7; tr. in De Bary, 489-90.

31 Baladhuri, Futuh al-Buldan, ed. De Goeje, 439/ed. al-Munajjid, 538.

32 Dvarasamudra: KF, 132, 135, 136. Tilang: NS, 116, 129. Nagarkot: SFS, 82. Cf. also DR, 46.

33 Taj, fols. 149a, 155a. TN, II, 79. 'Afif, 180, 264. IM, 63, 102. DA, fols. 32b, 33b-34a (tr. Rashid,

64, 65-6). TFS, 290, 586. SFS, 129, 167. FFS, 9, 10, 16 (tr. Roy,.456, 458, 462).

34 Fatawa-yi Firuz-Shahi, IOL Persian ms. 2987 (Ethe, no. 2564), fols. 410a, 412a, 414, 416a,

418b,419a.

~That the Indian polytheists who submitted to Islamic rule qualified, therefore, as 'protected

peoples' seems to have won acceptance among a fairly wide spectrum of the educated Muslim community

within the subcontinent. But the precise nature of the disabilities to be imposed on the infidel was a more

difficult matter. Early in the thirteenth century, Fakhr-i Mudabbir dedicated to Iltutmish his Adab al-Harb

wa'l-Shaja'a or Adab al-Muluk, a manual of statecraft for kings which is largely concerned with military

matters. In chapter 26 he reviews the principles and practice of Islamic governments regarding their non-

Muslim subjects, and lists the restrictions under which such people should live: their adornment (zayn), dress (jama) and deportment (nishast) are to be different from those of Muslims. He also lists the categories

of people who should pay the jizya, which includes Jews, Christians, Sabians, Zoroastrians (mugh) and

'idola-tors' (butparastan).35 This could be taken as evidence that Hindus were acceptable in the eyes of the

Ghurid conquerors of India as payers of the poll-tax; though it has been pointed out that the Addb al-Harb

is not a legal text and that it contains no explicit statement that Hindus are to be classed as dhimmis.36

Given the political circumstances prevailing in Muslim-ruled India in Fakhr-i Mudabbir's time,

how, where and upon whom was the jizya levied? We might expect some assistance in tackling the

problem from the accounts that have come down to us of the Muslim conquest of Sind in the eighth

century, and which could have served the thirteenth-century conquerors for precedent. Unfortunately, the

conquest of Sind preceded the emergence of a clear differentiation between the jizya and the kharaj. The earliest chronicles therefore afford us no real assistance. Baladhuri's Futuh al-Buldan says, in the course of

an otherwise full narrative, merely that Muhammad b. Qasim imposed the kharaj - either the land-tax or

just tribute - on the vanquished city of Alor.37 Not until a century later, in the caliphate of al-Mu'tasim (218-

27/833-42), do we find the Muslim governor of Sind taking jizya, in this case from the Jats.38 It is true that

a later source, the Chach-Nama, which purports to be a Persian translation, drawn up in Sind in 613/ 1216-

17, of an Arabic history of the Islamic conquest of the region, shows jizya being levied at the very outset.

Here Muhammad b. Qasim is alleged to have agreed that the inhabitants of Brahmanabad were to be

regarded as dhimmis and imposed on them a graduated tax in accordance with the tradition (sunan) of the

Prophet.39 The reliability of the Chach-Ndma is admittedly open to question. Dismissed by S. H. Hodivala

in 1939 as 'every whit as unhistorical as the similar lucubrations of Sanskrit poems and

35 AH, 404-5. 36 Hardy,'Growth of authority', 205-6.

37 Baladhuri, Futuh al-Buldan, ed. De Goeje, 439/ed. al-Munajjid, 538.

38 Ibid., ed. De Goeje, 445-6/ed. al-Munajjid, 544.

39 CN, 158-9. See also N. A. Baloch, 'Early advent and consolidation of Islam in the lands of

Pakistan', HI 3 (1980), 66.

~Rajput bards',40 the work has more recently been rehabilitated, and it is now believed to incorporate material from a lost Arabic historical tradition, most probably the ninth-century chronicle of al-

Mada'inl. Nevertheless, the data on the poll-tax are undoubtedly, anachronistic; and Dr Peter Hardy has

proposed that this kind of testimony in the Chach-Nama was designed to justify what had become standard

practice by the early thirteenth century. If he is right, this means at least that the jizya was being levied in

Sind on the very eve of the creation of the Delhi Sultanate.41

It is some measure of the problems surrounding the jizya that one of our most important sources,

Juzjani's Tabaqat, neglects even to mention it. The term surfaces fitfully in the sources for the thirteenth-

and early fourteenth-century Sultanate, but it evidently carries a variety of meanings.42

At times the usage is

bizarre, as when the jizya is allegedly taken from a rebel Muslim commander in the breakaway Deccan Sultanate or demanded from a Muslim mystic (darwish).43 The phrase kharaj-u jizya has contributed to the

confusion. Amir Khusraw uses it in a general sense, to mean tribute payable, for instance, by the enemy's

paiks, and such is clearly the sense also in Barani's Ta'rikh-i Firuz-Shahi.44 According to the Fatawa-yi Ja-

handari, Hindu rais levy the jizya and the kharaj from their own people.45 In none of these instances can we

discern the lineaments of the Islamic poll-tax.

Fakhr-i Mudabbir distinguishes two kinds of jizya. One is the tribute (gazid) agreed upon as the

price for a cessation of hostilities. The other type of jizya is the sum levied by a Muslim sovereign upon the

wealth - the houses, estates and moveables (khdna-u diya'-u 'aqar) - of individual infidels, and is gradated.

The annual rates given in the Adah al-Harb are precisely those specified in the Chach-Nama, namely forty-

eight silver dirhams for the richest, twenty-four for those of middling wealth, and twelve for the poorest.46

Ghaznawi, who here evidently follows Fakhr-i Mudabbir and gives the same figures, equates the first type of jizya with the kharaj-i muqasima (i.e. the land-tax proper). He also writes of two kinds ot tribute: that

rendered when a Muslim army has actually taken up its quarters in the infidel kingdom, which he classes

merely as booty (ghanima), and that offered prior to a Muslim invasion, which he calls jizya. At another

point he is prepared to class as jizya even money and gifts despatched

40 Hodivala, Studies, I, 83-4.

41 See P. Hardy, 'Is the Chach-nama intelligible to the historian as political theory?', in Khuhra

(ed.), Sind through the centuries, 116-17; idem, 'Djizya, iii. India', Enc Isl.2; also Yohanan Friedmann, The

origins and significance of the Chach-nama , in Friedmann (ed.). Islam in Asia, I, 23-37; Wink, Al-Hind, I,

192-6. 42 For what follows, see especially Hardy, 'Djizya'; I. Habib, 'Agrarian economy1, 67.

43 FS, 602 (tr. 888). Amir Hasan Dihlawi, Fawaidal-Fu'ad, 233.

44 RI, I, 33, and IV, 140. TFS, 291, 574. TMS, 147. This is what misled Lal (below, n.49).

45 FJ, 166.

46 AH, 404. CN, 158.

~intermittently by infidel princes.47 It is clearly jizya in the sense of tribute - a share of the land-revenue, surrendered by a Hindu rai - that was imposed, for instance, on King Rudradeva of Arangal in the

course of Malik Kafur's expedition in 710/1310-11 and again when Khusraw Khan invaded his dominions

in 718/1318.48

Certain historians have assumed that the jizya was levied on the subject Hindu population

throughout the era of the Delhi Sultanate.49 This seems unlikely. One relevant consideration is that the jizya

was a tax in lieu of military service and that - unlike Jews and (at least during this period) Christians in

other Islamic polities - Hindus, as we have seen, frequently fought in the ranks of Muslim armies; this

would have warranted the suspension of the tax in Muslim India.50 More relevantly, it has been argued that

the logistics of collection, involving enormous numbers of tax-payers and given the relatively

unsophisticated administrative apparatus of even a medieval Islamic state, must have presented an insuperable obstacle.51 On such grounds, it has seemed natural to conclude that the jizya did not exist as a

distinct tax but was subsumed within the kharaj or land-tax.52 Ghaznawi indeed envisaged that the two

might be consolidated as a single tax; though he urged that the respective proportions should be clearly

defined.53

Perhaps a distinction can be drawn between the Hindus of the rural areas and those living in the

Muslim-held towns and fortresses. In the case of the former, the jizya may have been perceived as forming

part of the land-tax or tribute rendered up by the chiefs;54 and this would make sense of the perplexing

remark cited above, from Barani's Fatawa, that Hindu kings (i.e. those tributary to the Muslim sovereign)

exacted kharaj and jizya from their own subjects. On the other hand, Barani attributes to Jalal al-Din Khalji a speech in which he refers to the paltry sums he accepts as sadaqa from the

47 DA, fols. 35a-36b (tr. Rashid, 67-8).

48 1311: KF, 111. 1318: NS, 84, 121. For jizya as tribute, see also QS, 35, 63; NS, 84, 121; FS,

275, 402 (tr. 450, 608-9); cf. also 35-7 (tr. 84-5), for a similar usage apropos of the Ghaznawid era, and 596

(tr. 879), where we are clearly dealing with the payment of two years' kharaj and the promise of future

tribute (sa-u baj). Likewise, IB, IV, 231 (tr. Gibb and Beckingham, 877), speaks of jizya being paid to the

Muslim ruler of Sumatra by his infidel neighbours.

49 Lal, History of the Khaljis, 184-5 (context is 'Ala' al-Din's reign). U. N. Day, Administrative

system of Dehli Sulanat (1206-1413 A.D.) (Allahabad, 1959), 106; in his The government of the Sultanate (New Delhi, 1972; 2nd edn 1993), 91-2, Day is more non-committal, though without implying that the tax

was introduced at a late stage. Zafarul Islam, 'Firuz Shah's attitude towards non-Muslims - a reappraisal', IC

64 (1990), part 4, 66, sees the jizya imposed in Firuz Shah's era as 'a revived impost'.

50 Aziz Ahmad, Studies in Islamic culture, 80-1: the context is the reign of the Mughal emperor

Akbar.

51 Hardy 'Djizya', and 'Authority of Muslim kings', 48.

52 Nizami, Some aspects of religion and politics, 314-15. See also Qureshi, Administration of the

Sultanate, 119.

53 DA, fol. 34a (tr. Rashid, 66).

54 This suggestion was made by Habibullah, Foundation, 250.

~Hindus of the capital. Sadaqa normally denotes the alms paid by Muslims, of course, and its use

here is a piece of irony, to suggest that the Muslim sovereign is in receipt of the unbeliever's charity. The

payment referred to may be the jizya.55 Conceivably the urban Hindu populace - artisans, members of

guilds, shopkeepers, and so on - who were in more direct contact with the Muslim fiscal authorities, had to

pay on an individual basis, i.e. a true poll-tax. It is surely no accident that Fakhr-i Mudabbir, in the passage

quoted earlier, speaks of the imposition of the canonical jizya in the context of the surrender of a town.56 As

far as I am aware, this solution to the problem has not been proposed before, but I offer it for what it is

worth (which, in the absence of strong textual backing, is not much). Whatever the truth, however, we first meet with incontrovertible evidence of the jizya as a discriminatory tax on individual non-Muslims only in

the reign of Firuz Shah (752-90/1351-88).

Firuz Shah's anonymous biographer assures us of that sultan's concern to impose no more than the

canonical taxes, incluDing 'the jizya of the Hindus' (jizya-yi Hunud).57 AccorDing to cAfif, he was the first

monarch to impose the jizya on the Brahmans, who had hitherto been exempt. (It is uncertain whether this

means that the jizya had actually been levied on other Hindu groups prior to Firuz Shah's time, or merely

that the sultan had himself excepted the Brahmans on a previous occasion, when imposing the tax on the

rest of Hindu society.) The Brahmans were scandalized and assembled outside his palace, threatening to

burn themselves to death. The sultan told them that they had better get on with it, since this was the only

way they would avoid payment - a somewhat cavalier response which gave no grounds for optimism. But a crisis was averted when the principal Hindu residents of the capital came forward with the offer to pay the

tax on the Brahmans' behalf. Firuz Shah in turn was ready to be more conciliatory, and taxed the Brahmans

at the lowest point on the scale, though using a tanga of different value.58 In his autobiography, Firuz Shah

mentions the jizya among the canonical sources of revenue permitted to a Muslim ruler, speaks of the

Hindus as submitting to the tax in return for protection of their property, and claims to have won over

countless Hindus to the true faith with an edict promising them release from the jizya if they would

convert.59

cAfif appears to specify that the rates he cites (ranging from twenty to forty tongas) applied in

Delhi. There is also evidence, however, from at least one province for the imposition of the jizya during Firuz Shah's reign. The tax is referred to twice in the correspondence of one of the sultan's officers, cAyn

al-Mulk Ibn Mahrti, governor of Sind. In the first case cAyn al-Mulk

55 TFS, 217. Cf. the remark in FJ, 167, that the infidels pay 'a few tangas by way of jizya''.

56 AH, 404.

57 SFS, 125.

58 cAfif, 382-4. For the significance of this passage, see Hodivala, Studies, I, 336-7.

59 FFS, 6, 9, 16-17 (tr. Roy, 453, 456, 462).

responds to protests at an increase in the jizya levied on Hindu shopkeepers.60 Later he refers to

the fact that the sultan had allocated to a military officer, as his stipend, the jizya paid by the peasants of a

certain district. The terms used show (a) that the tax was related to the protection of the dhimmi and (b) that

the owner (malik) of the land (in this case the qadis of Thanesar) had no claim upon it.61 This excludes any

possibility that we are dealing with the orDinary land-tax. The balance of the evidence, consequently, is

that in the latter half of the fourteenth century, if not before, the jizya was levied as a discriminatory tax on

non-Muslims;62 though even then it is difficult to see how such a measure could have been enforced outside

the principal centres of Muslim authority.

Latitude towards Hindu religious practice

There is little information in the sources about the attitude of the sultans towards Hindu religious

observance in general; and most of the evidence comes from the reign of Muhammad b. Tughluq, who was

hardly typical. Muhammad was notoriously interested in Hindu practices. He is charged by 'Isami,

admittedly no friend of the sultan, with attenDing the Hindu religious festival of Holi and (as Ibn Battuta

confirms) with frequenting the company of jogis.63 Ibn Battuta observes that the sultan's permission was

required for the ceremony of sati ('suttee'), the burning of a widow on her husband's funeral-pyre.64

Otherwise, the limited material at our disposal is concerned with the construction or repair of

Hindu temples. We saw earlier how in strict Islamic law it was not permissible for Hindus to build new

idol-temples or to restore those that had been destroyed. That this was being disregarded, however,- in cAla'

al-Din Khalji's reign is clear from Jain works which praise Alp Khan, his governor in Gujarat, for permitting the reconstruction of temples destroyed during the Muslim conquest.65 Fuller testimony is

provided by an inscription of 1326 from the Deccan. During the rebellion of Baha' al-Din Garshasp, the

governor of Kalyani, Ahmad Jajneri, was called away; and in the ensuing upheavals a Hindu temple at

Kalyani was damaged and the Siva linga was broken. Local Hindu notables, headed by the person in charge

of the management of the temple, therefore approached the governor on his return and sought his

permission for the repair of the temple and the resumption of the worship of the god. Ahmad Jajneri

consulted his secretary, whose name is certainly not a Muslim one, and granted permission, adDing that

since the worship of the god was a duty it

60 IM, 48.

61 Ibid., 62-3.

62 See also Zakir Husain, 'Some original Tughluq documents and their significance', PIHC 50.

(Gorakhpur 1989) (Delhi, 1990), 222.

63 FS, 515 (tr. 765). IB, IV, 36, 38-9 (tr. Gibb and Beckingham, 788, 789-90).

64 Ibid., III, 137 (tr. Gibb, 614).

65 Misra, Rise of Muslim power, 68-9.

was right that the petitioners should pursue it. This testimony has been cited as indicating the

existence of a striking degree of tolerance.66 But it comes, of course, like the material relating to Alp Khan,

from a region that had only recently come under direct Muslim rule, and one where the sultan's authority

must have been highly precarious. It does not tell us what might have been the response had a similar

situation arisen in a core territory like Hariyana.

We have, in fact, other evidence for the latitude enjoyed by local Hindu religious authorities at this

time. Ibn Battuta writes of an embassy from 'the king of China' (i.e. the Mongol Yuan emperor) to

Muhammad b. Tughluq, requesting permission for the reconstruction of a temple in the Sambhal region

which had allegedly been sacked by a Muslim army. The envoys were told in reply that permission to

restore such temples within Muslim territory could be given only to those who paid the jizya (and not, in other words, to infidels resident in the Dar al-Harb).67 The authenticity of this embassy is questionable, and

it is in any case conceivable that Ibn Battuta, as an outsider, misconstrued the state of affairs. But the

likelihood is that his testimony regarDing the reconstruction of idol-temples is reliable. For indeed Firuz

Shah claims that prior to his accession new temples had been built in Delhi and its environs - contrary, of

course, to the Shari'a.68 Of this there survives, unfortunately, only meagre direct evidence. An inscription

shows that a new temple was built at Revasa, in the Nagawr region, in 1326;69 a fragment of a bilingual

inscription, in Sanskrit and Persian, of uncertain date but very probably from the Sultanate period, records

the purchase of twelve bighas of land near the Qil'a Kuhna in Delhi itself and the erection of the temple of

Sri Krishna Bhagwan.70

It is not beyond the bounds of possibility that such foundations received endowments from the Muslim authorities. Alp Khan is said to have made a donation towards the repair of Jain temples in

Gujarat;71 but otherwise the large numbers of documents that attest the conferment of land and tax

exemptions by Muslim kings on Brahmans, Jains, jogis and Parsis and on temples to Siva and Visnu tend to

originate from the Mughal emperors and their contemporaries in the successor-states to the Delhi Sultanate.

Although many of these confer new revenues, some are clearly renewals or extensions of grants made by

Muslim rulers of an earlier era.72 There are signs of donations of tax-free land (madad-i macash) to

Brahmans during at

66 P. B. Desai, 'Kalyana inscription of Sultan Muhammad', 165-70. See HN,503; also W. H.

Siddiqi, 'Religious tolerance as gleaned from medieval inscriptions', in PSMI, 54, where the governor is

mistakenly identified, however, as Ahmad-i Ayaz, the sultan's future wazir.

67 IB, IV, 1-2 (tr. Gibb and Beckingham, 773).

68 FFS, 9-10 (tr. Roy, 456-8).

69 Progress Report of the ASI (Western Circle) (1909-10), 52, cited in Welch and Crane, The

Tughluqs: master-builders', 160 n.l1.

70 ASIR (1909-10), 131.

71 Misra, Rise of Muslim power, 69.

72 Ernst, Eternal garden, 48-9.

least the period of the Lodi sultans, and it may be that such grants were made by their predecessors

also. Professor Siddiqui sees this as the sultans' response to the need to bring 'the countryside with its

influential chiefs under effective control'.73

Generally speaking, Firuz Shah's policies weighed more heavily upon the subject Hindu

population than those of his predecessors. cAfif tells how the sultan burned a Brahman at the palace gates.74

Firuz Shah himself, in describing the new temples that had arisen under his predecessors, claims to have set

about destroying them and replacing them with mosques, and in one instance to have repopulated a township with Muslim settlers.75 Two points must be emphasized here. Firstly, these were all new edifices:

there was no' question of destroying temples and shrines which had already existed before the Islamic

conquest and whose devotees lived peacefully under Islamic government. And secondly, these events all

transpired in the vicinity of Delhi. The sultan's writ would hardly have extended to enforcing such a policy

over a wider radius. This is clear from his conduct in sparing the idol of Jawalamukhi at Nagarkot, a step

that was, in fact, perfectly in keeping with the policy of earlier Muslim rulers. But it was difficult to

reconcile with his iconoclastic image, and gave rise to rumours spread by 'certain infidels' - and which 'Afif

was at pains to refute - that the sultan had paid his respects to the idol and unfurled a chatr over its head.76

Nevertheless, whatever qualifications are made as to scale, it cannot be denied that Firuz Shah's reign

witnessed a reaction against previous regimes. After his death, the Hindus' situation may have deteriorated

further in certain regions: a temple at Ketlai, in the Gurga'un district, was destroyed in 795/1392 and

replaced by a mosque.77

Hindu-Muslim relations: an assessment

In relating military encounters with the Hindu, the narrative sources abound in unflattering, if

conventional, allusions: one of the most frequent is the description of the enemy as 'crow-faced (zagh-

chihra) ,78 But at times a more neutral tone is heard. Juzjani hoped that the qualities of King Laksmanasena

of Bengal, who had gained a reputation for justice and

73 I. H. Siddiqui, 'Wajh-i Macash grants under the Afghan kings (1451-1555)', MIM 2 (1972), 21,

36. For the Mughal era, see B. N. Goswamy and J. S. Grewal (eds.), The Mughals and the Jogis of Jakhbar

(Simla, 1967): evidence from the regional sultanates in the pre-Mughal period is given ibid., 20. 74 cAfif, 379-81.

75 FFS, 9-10 (tr. Roy, 456-8).

76 cAfif, 186-7. The sparing of the idol is mentioned briefly in SFS, 83.

77 ARIE (1963-4), 146 (no. D286).

78 Annemarie Schimmel, 'Turk and Hindu: a poetical image and its application to historical fact', in

Speros Vryonis, Jr (ed.), Islam and cultural change in the Middle Ages (Wiesbaden, 1975), 107-26: see esp.

109-16.

generosity, would earn him alleviation of his torments in Hell.79 When Toghril Khan Yuzbeg, the

muqtac of Lakhnawti, repudiated the sovereignty of Sultan Nasir al-Din Mahmud (c. 652/1254), says the

same author, his action incurred the disapproval of Hindus and Muslims alike within the Sultanate.80 Amir

Khusraw has been singled out as one who in his Nuh Sipihr drew attention to the kinship between certain

Hindu religious beliefs and those of Islam.81 At the tragic death of Balaban's son Muhammad in battle with

the Mongols, wrote Khusraw, 'the Hindu lost his blackness and the Turk his whiteness'. Literary device this

may be; but the implication is that Khusraw, like Juzjani, thought the Hindu's view could be taken on

board. And it seems that the Hindu merited a place in the divine dispensation when compared with other

pagans. The Mongols, who were viewed as harbingers of the last things (p. 113 above), were described in

much more opprobrious terms than were the Hindus in Indo-Muslim writings. Khusraw derived some satisfaction from the fact that Providence had used cAla' al-Din's infidel general Nanak to defeat the infidel

Mongols.82 Legal texts of the fourteenth century reveal a concern about the relations of orDinary Muslims

with the Hindu population. The Fatawa-yi Firuz-Shahl, in particular, pronounces on the proper conduct of

social intercourse with Hindus, the right treatment of Hindu parents by a Muslim son, the equal rights of

Hindu and Muslim creditors, and so on.83

All this might seem to stand in sharp contrast with the tone adopted by Barani. A theme that recurs

frequently in Barani's writings is that the infidels must on no account be allowed to live in ease and

affluence. In the preface to his life of the Prophet, Na't-i Muhammadi, he cites a dispute that allegedly took

place at the court of Iltutmish a century and a quarter earlier. When the 'ulama' declared that the Hindus had no right to be treated as 'Peoples of the Book' and should be given only the choice between death and Islam,

the sultan's wazir Junaydi is said to have agreed with them. But, he continued, such a course would be

highly impolitic, given that the Muslims were still few in number, and its implementation should be

deferred until they were in a stronger position. The culama' thereupon insisted that the sultan should at least

refrain from treating Hindus with honour or permitting idolatry in the capital. But it was because of this

failure to slaughter the Hindus, says Barani, that polytheism had taken root.84 This is echoed in another

hypothetical conversation from the same

79 TN, I, 425 (tr. 555-6).

80 Ibid., II, 32 (tr. 764).

81 Yohanan Friedmann, 'Medieval Muslim views of Indian religions', JAOS 95 (1975), 216-17.

82 WH, IOL Persian ms. 412, fol. 135a. KF, 38. DR, 61. The comparison with the treatment of the

Mongols is made in Hardy, 'Growth of authority', 193.

83 Z. Islam, 'Fatawa Firuz Shahi as a source', 105-7.

84 Barani, Nact-i Muhammadi, RRL Pers. ms. 1295, fols. 195b-196a; tr. in S. Nurul Hasan,

'Sahifa-i-Nact-i-Muhammadi of Zia-ud-Din Barani', MIQ 1 (1950), 101-3 (Pers. text at 104-5).

era which is found in Barani's Ta'rikh. The sign that a ruler protects the true faith, Sayyid Nur al-Din Ghaznawi tells Iltutmish, is that when he espies a Hindu his face grows red and he wants to bury him

alive. If the polytheists are so numerous that the Muslim ruler cannot possibly eradicate them, then at the

very least he should strive to insult them and bring disgrace, dishonour and ignominy upon them.85 And the

same theme recurs in a speech attributed to one of cAla' al-Din Khalji's advisers, Qadi Mughith al-Din of

Bhayana. In answer to a question from the sultan about the status of the Hindu in the Sharia as regards

taxation, the qadi asserts that when the tax-collector demands silver from the Hindu he should mildly,

humbly and respectfully hand over gold; and if the tax-collector throws dirt in his mouth, he should open

his mouth to receive it.86

Such views are commonly encountered in polemical writing against the infidel in different parts of

the Islamic world at different times.87 But there are other notions that are peculiar to Barani himself. For

Barani, it is one of the primary duties of Muslim kings to redeem the inherently sinful and evil nature of kingship by rooting out paganism, polytheism and idolatry.88 The Hindus are the worst enemies of God and

his Prophet.89 Indeed, the Prophet had commanded that they were to be looted and enslaved or killed.90 The

Brahmans in particular, who are the leaders and instigators of idolatry, should be massacred.91 Only the

Hanafi school of law allows that the Hindus qualify to pay the jizya; the founders of all the other schools

insist that the sole choice to be offered to Hindus is Islam or death.92 Much of this is blatantly unhistorical.

In his Fatawa-yi Jahandari, which masquerades as a political testament from Mahmud of Ghazna, Barani

makes further statements that are equally dubious. Had Mahmud invaded India just once more, he would

have slaughtered all the Brahmans and beheaded 200,000 or 300,000 Hindu chiefs (an intriguing

demographic statistic).93 Mahmud is said to have confided to Qadir Khan, the Qarakhanid ruler of

Kashghar, his fear that on the Day of Judgement he would be asked why he had not killed the Brahmans94 -

and this when the real Mahmud had been condemned by contemporary Muslim chroniclers for employing infidel Hindu troops against fellow Muslims during his campaigns in Persia.95

Yet Barani's antipathy towards the infidel Hindu can be overstated. For all his railings, he is

evidently aware that the contradiction between the demands of orthodox Islam and the situation in India

cannot be resolved.96 Moreover, he does permit himself the occasional neutral reference to the

85 7FS,41-2.

86

Ibid., 290.

87 For examples, see Lewis, Jews of Islam, 32ff.

88 TFS, 41. See generally Hardy, 'Oratio recta, 319.

89 TFS, 42, 290; Nact-i Muhammadi, fols. 195b-196a, tr. Nurul Hasan, 102 (Pers. text 104, 105).

90 TFS, 290-1.

91 Ibid., 42. FJ, 165.

92 TFS, 291. Cf. also FJ, 18.

93 Ibid.

94 Ibid., 230.

95 Bosworth, Ghaznavids, 89, 110.

96 I. Habib, 'Barani's theory', 111-13.

Hindus. He thinks it worth mentioning that during the famine of 1291 in Delhi Hindus came in

groups of twenty or thirty to throw themselves into the Yamuna.97 And he observes that Hindus as well as

Muslims prayed for Muhammad b. Tughluq on his accession and rejoiced at the advent of Firuz Shah in 752/1351 and at his safe return from his first Bengal expedition a few years later.98 These remarks suggest

that, although Barani would not for one second have considered his Ta'rikh-i Firuz-Shahi as a history of the

entire population of the Sultanate, he did not, even so, deem the non-Muslim section of that population to

be totally beneath the historian's notice. Perhaps the most arresting indication of a different frame of mind

from the polemics outlined above is his statement that he is now ready to contemplate the life of a

Brahman.99 In other words, he does not at this juncture, as one might expect, hold up as his goal the calling

of a sufi. It is possible that in his old age, and confronted by a sharp decline in his material condition,

Barani found something to commend in the degree of self-abnegation attained by certain leading exponents

of the rival faith.

Whatever the case, Barani's fulminations against gentle treatment of the non-Muslim must be seen

for what they are. Not merely were his writings drafted largely from memory by a man advanced in age; not merely do they exhibit at times a lamentable ignorance of history; they are also the product of a courtier

who had fallen from favour after the death of Muhammad b. Tughluq and who bitterly resented his change

of fortune. Barani wrote, moreover, as the representative of a family that had served Balaban's officers and cAla' al-Din Khalji. His paternal ancestors may have been Turkish; more probably they were of Persian

stock. His father's mother was of the illustrious lineage of the sayyids of Kaithal.100 Barani accordingly

prides himself on his high birth and has no time for those of lowly origin. Significantly, his list of

Muhammad b. Tughluq's lowborn servitors includes not only Hindus but also those who, to judge from

their names, had embraced Islam. Professor Irfan Habib has in fact pointed out that, unlike cIsami, Barani

did not attack the sultan for his favour towards Hindus and that his objection to the promotion of these men

- Hindu and Muslim alike - was based above all on their humble origins.101 In some measure, certainly,

Barani's assertions about the status of the infidel are part of his more general indictment that men of low birth had benefited (and more than he himself had done) from his late master's patronage. But in fact

Barani, in his denial that the essentials of Islam can be implanted in the minds of Indian converts, seems to

share the prejudice of cIsamI, for whom 'a Hindu ghulam will flee in the end, though he attain the rank of

chief sadr'.102

97 TFS, 212.

98 Ibid., 457, 547, 596. I. Habib, 'Barani's theory', 113.

99 TFS, 200.

100., Ibid., 350.

101 Ibid., 504-5; and cf. FS, 486, 515 (tr. 728, 765). I. Habib, 'Barani's theory', 110.

102 FS, 552 (my translation); cf. also 370 (tr. 571). FJ, 105.

It may be possible to identify more general causes of antagonism towards the infidel on the part of

Muslim writers and rulers. One may have been a fear of apostasy on the part of ordinary Muslims. Here a

parallel offers itself with Western Christian attitudes towards Islam in the Middle Ages. A common theme

in Christian polemical writings on Islam and the Muslims is the low standard of sexual morality

encouraged by the rival faith, whether on the level of polygamy, concubinage, ease of divorce, and ideas about paradise on the one hand or, on the other, the charge of dark and unnatural practices which would

have been harder to substantiate but was no less sinister for being left vague.103 In harping on such matters,

Christian authors unconsciously testified to the attractive force of a religion that in their eyes was

calculated to appeal to the sensual and the self-indulgent. We can, I suggest, detect a parallel phenomenon

within Muslim circles in India. Beneath the surface of the political events on which our narrative sources

focus lay a substratum of everyday Muslim-Hindu intercourse. Both Muslim and Hindu musicians

performed at the celebrations for the marriage of Prince Khidr Khan.104 Hindus are said to have mingled

with Muslims in the crowds that gathered to celebrate the festival of Barat (14 Shaban).105 Hindus and

Muslims sometimes rubbed shoulders at the entrance to the hospices (khanaqahs) of sufi shaykhs.106

Conversely, many Muslims attached themselves to a group of jogis at Khajuraho (Kajarra) in order to

acquire their skills.107 Muhammad b. Tughluq was only the most eminent figure to share in Hindu festivities: Ibn Battuta saw Muslims in the throng accompanying a widow on her way to be burned, and

both Firuz Shah and his biographer accuse orDinary Muslims of participating in Hindu religious rites.108

The Brahman executed on Firuz Shah's orders was charged not merely with hosting idolatrous ceremonies

in his house that were attended by Muslims, but also with inducing a Muslim woman to apostatize.109

Muslims had been known, moreover, to flee into infidel territory. Some were prominent nobles

like those implicated in the conspiracy during the Tilang campaign of 721/1321-2; but there were always

renegades like the Muslims, Ibn Battuta claims to have met when captured by Hindus near Jalali.110 For just

as Hindu troops fought under the banner of Muslim sultans, so did Muslim soldiers fight for infidel rulers -

whether it was the Muslims in the army of the Pandya king of Macbar in 710/1310-11 or, some three

decades later, the 20,000 Muslims, 'rascals, criminals and runaway slaves', who are reported in the service

of the Hoysala king Ballala III, or 103 Norman Daniel, Islam and the West: the making of an image (Edinburgh, 1960), 135-61.

104 TMS, 79.

105 cAfif, 366.

106 Troll (ed.) Muslim shrines, 7, 14.

107 IB, IV, 40 (tr. Gibb and Beckingham, 790).

108 Ibid., III, 137 (tr. Gibb, 614). cAfif, 380. FFS, 9-10 (tr. Roy, 457). See also Ernst, Eternal

garden, 27 and 289 n. 107.

109

cAfif, 379-81.

110. IB, IV, 10-11 (tr. Gibb and Beckingham, 777).

the 'worthless' (nabakar) Muslims under the banner of Narayan when he marched against the

founder of the Bahmanid dynasty.111

In the circumstances, there was perhaps an uneasy sense that the

Muslim minority in an overwhelmingly pagan land might be seduced into infidelity. Mihrabi's Hujjat al-Hind, a treatise that has survived from the end of the fourteenth century, aims at countering just such

apostasy in the countryside.112

The sultans were undoubtedly also a prey to pressures of a different sort. Historians of the Islamic

world, notably Professor Bernard Lewis, have demonstrated that Muslim rulers can never be treated as

monolithic in their approach to their non-Muslim subjects. Their policies fluctuated according to

circumstance: an external military threat posed by the co-religionists of the subject group in question; the

need on the part of the ruler to reassure the Muslim population if it was felt that other confessional groups

had benefited from excessive leniency or favour; or simply the desire of a new sultan to buttress an

authority that was of doubtful legitimacy with the support of orthodox jurists and preachers.113

Circumstances of this order cannot be ignored either in an analysis of Hindu-Muslim relations within the

Delhi Sultanate. Take, for example, the patronage of Hindus by Muhammad b. Tughluq. Muhammad also made greater efforts than any other Delhi ruler to attract into his service Muslims from every part of the

Islamic world (pp. 184-5, 233-4). The paradox here is more apparent than real. These were two arms of a

policy which aimed at creating a counterweight to the Indian Muslim nobility, since the Tughluqid dynasty

had come to power only a few years previously in the teeth of determined Indian Muslim opposition. We

should also bear in mind that a dramatic extension of Muslim power had occurred. The wars of the past

three decades had eliminated most of those major independent Hindu kingdoms which might have

presented a competing focus of allegiance. It is instructive in this respect to compare the position of

Christians in, say, Egypt or Anatolia, where the Muslims could not afford to forget their relations with

interested foreign powers. The Delhi Sultan was able to promote Hindu servitors as he did, or patronize

Hindu religious establishments when it suited him, precisely because India contained no rival imperium like

Byzantium or the states of Catholic Europe.114

Firuz Shah, in turn, may well have been a more orthodox and pious figure than his late cousin; but

extraneous factors also surely underlay his policies. The sultan's accession had not gone unchallenged; and

he was

111 KF, 149, and DR, 72, for the Pandya army; IB, IV, 195-6 (tr. Gibb and Beckingham, 861), for

that of Ballala III; FS, 592 (tr. 873), for Narayan. Other examples in Bouchon, 'Quelques aspects', 30.

112 Peter Hardy, 'Islam and Muslims in South Asia', in Raphael Israeli (ed.), The crescent in the

east: Islam in Asia Major (London, 1982), 43.

113 Lewis, Jews of Islam, 32-61 passim. 114 A point well made by Ernst, Eternal garden, 50-1.

clearly conscious, moreover, of a need to distance himself from the extravagances of Muhammad,

who had clashed with the Islamic 'religious establishment' and executed not a few of its members. It is also

important, in this connection, that policies towards dhimmis are not seen in isolation from other measures.

They were often linked with attempts to suppress heterodox Muslims or with the abolition of uncanonical

taxes, both actions for which Firuz Shah took care to be known and for which he is lauded by his

biographer cAfif.115 A clampdown on the dhimmi, rather than being seen as an end in itself, has to be

viewed as part of a broader policy.

I have tried, in examining the all too meagre evidence at our disposal, to offer a perspective on

Hindu-Muslim relations which might indicate that conventional formulations will not do. It is impossible to

tell how far Barani typified, in his attitude towards the Hindus, the class of Muslim literati; and even his

outlook was a curious amalgam. But his very stridency at times suggests, and the evidence of other sources

confirms, that there was a wide gulf between Islamic law and the inclinations and practices of rulers and the

military class. The sultans were faced with a situation which had not confronted Muslim rulers elsewhere

for a few centuries. 'Do you not see,' Barani makes Jalal al-Din Khaljl ask,

'how every day the Hindus ... pass beneath my palace, beating their drums and blowing on their

conches, and make their way to the Yamuna to practise their idolatry, and how they fulfil the requirements of polytheism and paganism before my eyes ... - I, who have myself called the ruler of the Muslims and

sovereign of Islam? ... Shame on me and on my kingship ... that I permit my name to be recited from the

pulpit every Friday and the preachers with their lying tongues to proclaim me as the defender of Islam,

when under my rule the enemies of God and of the faith, in my sight and in my capital, live in affluence

and ease and surrounded by a thousand luxuries ... and strut about among Muslims and openly practise

idolatry '116.

The speech is apocryphal like all the others, but the words Barani has put into the sultan's mouth

have an authentic ring: they describe not what ought to be, but what is. The decisions made by the sultans,

of how to comport themselves towards the infidels who represented the majority of their subjects, were

informed by more complex considerations than we have often supposed. We surely have to begin with the

presumption that within their own dominions, for some of the time, they managed to approach the problem, not as iconoclastic holy warriors, but with a degree of delicacy. Perhaps for them the paramount distinction

was not that between Muslim and Hindu (important as that may have been) but between peaceful subject

and agent of government on the one hand and troublemaker and rebel on the other.

115 FFS, 5-6 (tr. Roy, 453-4). cAfif, 373-9. See generally Z. Islam, 'Firuz Shah's attitude'.

116 TFS,216-17.

CHAPTER 15

Stasis and decline: Firuz Shah and his successors

The contracted Sultanate

To judge from the remarks of the 'official' chronicler, it was a matter of some pride at Firuz Shah's

court that Ilyas Shah, the upstart sultan of Bengal, had begun his career as merely the servant of an officer

of one of Muhammad b. Tughluq's servitors (see above, p. 267).l He was, moreover, a tyrant, and Firuz

Shah could not be impervious to the appeals of his wretched subjects for deliverance.2 So too, the patents

which the new sultan received from the cAbbasid Caliph - judging, again, by the treatment they are

accorded in the Sirat - had an important share in buttressing the exclusive legitimacy of his government.

The Egyptian shaykh al-shuyukh Rukn al-Din al-Malati, who had brought a diploma for Muhammad from

al-Hakim, left India early in Firuz Shah's reign, arriving back in Cairo early in 754/1353 after an absence of

nearly ten years.3 But that same year an embassy arrived in Delhi from the Caliph al-Muctadid bi'llah, bringing Firuz Shah a mandate (manshur) for the government of India and conferring on him the titles Sayf

al-Khilafat ('Sword of the Caliphate') and Qasim Amir al-Mu'minin ('Partner of the Commander of the

Faithful'). In 764/1362-3 the next caliph, al-Mutawakkil ila'llah, despatched another mission to Delhi with a

mandate in which Firuz Shah was addressed as Sayyid al-Salatin ('Lord of Sultans') and declared to be the

caliph's wall. Similar embassies followed in 766/1364-5 and, according to the Sirat, each year thereafter.4

Al-Mutawakkil was at pains to stress that to obey the sultan was to obey

1 SFS, 47.

2 Farman of Firuz Shah reproduced in IM, 16; tr. in Maulavi cAbdu'l Wali, 'Life and letters of

Malik cAynu'l-Mulk Mahru and side-lights on Firuz Shah's expeditions to Lakhnauti and Jajnagar', JASB ns 19 (1923), 279. SFS, 34 (tr. Basu, JBORS 23 [1937], 111). cAfif, 137-40, 143.

3 al-Maqrizi, Suluk, II, part 3, 887. His figure of ten years and nine months must be incorrect

(above, p. 272).

4 SFS, 282-5 (tr. Rashid, 'Firuz Shah's investiture', 70-1). cAfif, 274-6, mentions only one caliphal

embassy to Firuz Shah, from al-Muctadid. TMS, 126, has the first one (but from al-Hakim) in 757/1356; see

also 127.

the caliph himself; that the sultan was empowered to wage jihad against rebels; and that neither he nor his two predecessors had issued a mandate to any Indian ruler other than the Delhi sovereign.5 The

caliph thus constituted Firuz Shah his intermediary in dealing with other Muslim princes in the

subcontinent. The territories listed in al-Muctadid's diploma included not merely Bengal, Ma'bar, Tilang,

Deogir, Kawlam (Quilon), Hinawr, Bakanawr and the rest of the coastal regions (sawahil-i bahr), but also

'the island of Sarandib' (Ceylon), 'the Jawat' (Greater and Lesser Java), the Qarachil mountains, 'the Afghan

territory (hudud-i Afghaniyya) and its mountains as far as Kashmir, and Zawulistan as far as the frontiers of

the Turks and Ma wara' al-Nahr'.6 This was, at best, a programme for future conquest (and reconquest).

In the early part of the reign, cAfif learned from his parents, there had been a period of seven years

in which Firuz Shah spent a total of merely thirteen days in Delhi: each time he returned from some

protracted campaign, he was off again almost as soon as he entered the capital.7 This restlessness may have

sprung from a consciousness of the grave territorial losses inflicted on the Sultanate in the time of Muhammad b. Tughluq. The welcome news of Taghai's death at the hands of loyalist commanders in

Gujarat had reached Firuz Shah on the day of Khwaja Jahan's submission.8 But the rebel's Sumra allies

were still at large; and we are given the impression that the sultan was deeply sensitive to the humiliations

suffered by his cousin in Sind and determined on vengeance.9 Nor could he remain oblivious of the loss of

provinces south of the Vindhyas.

The reality, of course, was that Firuz Shah was in no position to retrieve the territories lost in his

predecessor's reign. Ma'bar had to be consigned to oblivion. Envoys from Macbar who waited on the sultan

following his return from the Thatta campaign claimed that their ruler had been defeated and put to death

by Bukka, the king of Vijayanagara, and that the Muslims were in desperate straits. Firuz Shah temporized,

observing petulantly that when he had sent them a farman at his accession the people of Macbar had failed to acknowledge his authority and now implored his aid because they were hard pressed; he would march

south once his troops were rested.10 At another point, we are told, the sultan set off hunting in the direction

of

5 SFS, 283-4 (tr. Rashid, 70-1).

6 Ibid., 283 (tr. Rashid, 70, omits Deogir inter alia). Bakanawr is the Fakanur of IB, IV, 78-9 (tr.

Gibb and Beckingham, 808). For the two Javas, i.e. Sumatra and Java proper (sometimes called Mul Jawa

in Islamic sources), see ibid., TV, 228-47 (tr. Gibb and Beckingham, 876-84); Pelliot, Notes on Marco

Polo, 755-8.

7 cAfif, 399. 8 SFS, 19, 27-8 (tr. Basu, JBORS 23 [1937], 105-6).

9 cAfif, 191-2.

10 Ibid., 261-3. Hodivala, Studies, I, 326-7, proposes that the dead ruler, called a kinsman of Hasan

Gangu, was Fakhr al-Din Mubarak Shah, the penultimate sultan of Macbar. For the rulers of Macbar during

this period, see S. Abdul Qadir Husaini, 'The Sultanat of Madura', in HN, 1023-5.

Dawlatabad, but turned back at Bhayana in view of 'the interests of the kingdom' and headed an

expedition to Nagarkot instead.11 At the time of his second attack on Thatta, Bahram Khan Mazandarani, son-in-law of Hasan Gangu, in the course of a struggle with the latter's son, sent word inviting Firuz Shah

to come south and take over at Dawlatabad; but the sultan, who was currently refitting his army in Gujarat,

decided to give priority to Thatta and the opportunity was lost.12 Subsequently, Firuz Shah announced his

intention of marching on Dawlatabad, i.e. to overthrow the Bahmanids, only to be dissuaded by the wazir

Khan Jahan on the grounds that it was unacceptable to make war on Muslims.13 Nevertheless, the

Bahmanids may have continued to fear an invasion from the north, since according to cAfif the effective

administration of his governor Zafar Khan (II) in Gujarat (in the late 1370s) caused trembling (larza) in

Dawlatabad.14 Those military enterprises that Firuz Shah did embark on have scarcely commended him to

modern historians. Dr Banerjee characterizes him as 'not even a mediocre military leader'; for Professor

Saksena he was 'not the stuff conquerors are made of; and Professor Riazul Islam declares that 'as a general he was thoroughly incompetent'.15 These views find ample support from the sultan's operations in Bengal.16

Large armies were mustered in 754/1353 against Shams al-Din Ilyas Shah, who had twice encroached upon

the Delhi Sultan's territory, and in 760/1359 against his son and successor Sikandar. During both invasions,

the Delhi forces holed up the enemy in the island fortress of Ikdala,17 only then to abandon the campaign.

On the first occasion, Firuz Shah managed to defeat Ilyas's forces in a pitched battle and to occupy the town

of Ikdala but, moved by the lamentations of Muslim women in the citadel, he rejected his generals' advice

to storm it.18 In the campaign against Sikandar, Firuz Shah was bought off with gifts and made peace on

condition that his client Zafar Khan was installed as ruler of Sunarga'un; but the affair drifted into farce

when Zafar Khan, conscious that he lacked any real power-base in Sunarga'un, opted instead to return with

the sultan to Delhi.19 As the Delhi army retired via Jajnagar, it lost its way, and it was six months before the

sultan rejoined his heavy baggage at Kara.20 In the wake of each campaign, the Bengal ruler sent elephants

and other gifts to Flruz Shah, and cAfif

11 cAfif, 185-6.

12 Ibid., 224-5. See Hodivala, Studies, I, 322, for the chronology.

13 cAfif, 263-6.

14 Ibid., 499.

15 Banerjee, History of Firuz Shah, 28; see also 26, 32-3. Saksena, in HN, 582. Riazul Islam,

'Flruz Shah Tughluk', Enc.Isl2, II, 924. 16 For Firuz Shah's Bengal campaigns, see Banerjee, History of Firuz Shah, 28-36, and HN, 582-

5,589-91.

17 For the location of Ikdala, now a village in the Dinajpur district and situated about 23 m. N. of

Pandua and 42 m. N. of Lakhnawti, see Hodivala, Studies, I, 311-12.

18 cAfif, 118-19.

19 Ibid., 156-8, 162.

20 Ibid., 172-3. Saksena (in HN, 593) discounts this story, which is found in no other source.

claims that the sultan remained on friendly terms with Sikandar, exchanging presents annually

with him until his own death.21

It was not until c. 767/1365-6 that the sultan moved against the Indus delta, now the domain of the

two Samma princes, cAla' al-Din Jawna, who bore the title of jam, and his nephew Banbhina.22 Their

villainy (fasad), according to the Sirat, had lasted for a generation.23 Ibn Mahru's correspondence shows

that Banbhina had raided Gujarat and had attacked the Panjab with Mongol assistance, probably during

Firuz Shah's second absence in Bengal. The government was endeavouring to bolster the position of the

Sumra prince, Hammir Duda, who was likewise under threat from the Sammas.24 The Sirat, composed only

a few years later, would have us believe that the Jam and his nephew were forced to sue for peace and that Firuz Shah generously granted them terms, but other authors are less sanguine. In cAfif 's version, the tone

of which is echoed by Sirhindi, Firuz Shah's operations had been a failure. An epidemic had carried off

three-quarters of his horses, and rising grain prices caused a famine among the troops, so that after a few

weeks of skirmishing he withdrew into Gujarat to refit his army. During the retreat the entire fleet fell into

the hands of the enemy, and en route for Gujarat the guides led the Delhi army into the Rann of Kachh, a

saline wilderness where the troops suffered dreadfully. When Firuz Shah finally reached Gujarat, its

governor was dismissed for his failure to furnish provisions and fresh troops.25 But at length the sultan was

able to return to Sind, and timed his arrival so as to appropriate the harvest on which the Samma forces had

relied. Nevertheless, they offered a desperate resistance, and a prolonged conflict was averted only by the

intervention of the local saint Jalal al-Din Bukhari, which may have been as timely for the Delhi army as it was for the Sammas. The Jam and Banbhina submitted and the sultan took them back with him to attend his

court at Delhi, leaving the Jam's son and Banbhina's brother Tamachi as joint rulers to represent his

interests.26 Trouble continued from this quarter, however, and Firuz Shah later had to send the Jam to

suppress a revolt by Tamachi.

21 cAfif, 161. TMS, 126, 128. TFS, 597, also refers briefly to gifts from Ilyas.

22 For the operations in the Indus delta, see generally Banerjee, History of Firuz Shah, 36-40; HN,

595-9. The date is discussed in Hodivala, Studies, I, 322, on the basis of the statement in cAfif, 191, that

four full years had elapsed since Firuz Shah's return from Jajnagar (dated 762/1361 in TMS, 130).

23 SFS, 84. 24 IM, 100-3, 230-5; tr. in Riazul Islam, 'Rise of the Sammas', 361-2, 368. N. B. Ray. 'Interesting

side-light on Firuz Shah Tughlaq's expedition to Tatta', JASB, Letters, 3rd series, 4 (1936), 285-92, gives

the first of these letters in full, but Hammir Doda's name is garbled so that the translation omits all mention

of him: see ibid., 286 n.2, for the date of this letter.

25 SFS, 86-7. cAfif, 200, 201, 207-8, for the epidemic and the famine; 203-5, 220, for the decision

to refit the army in Gujarat; 207 for the loss of the fleet; 208-19 for the Rann; 219-20 for the governor's

dismissal. TMS, 131. Bihamadkhani, fol. 410b (tr. Zaki, 9).

26 Riazul Islam, 'Rise of the Sammas', 377-9, citing two versions of the saint's table-talk, the

Malfuzat-i Makhdum-i Jahaniyan and the Siraj al-Hidaya.

The Sammas appear eventually also to have vanquished the Sumra prince Hammir Duda, who is

found in exile in Gujarat before the end of Firuz Shah's reign.27

Nor were Firuz Shah's military triumphs over Hindu powers on such a scale as to redeem his

reputation. Every year, says cAfif in one of his encomiastic passages, the people of the Dar al-Harb were

raided and plundered.28 The purpose, as so often in the past, was to obtain overdue tribute. En route for

Bengal in 754/1353, the sultan had taken the opportunity to assert his authority over the local Hindu chiefs

when he reached Awadh and to exact arrears of tribute from the rais of Kharonsa and Gorakhpur, of whom

the latter, Sirhindi tells us, handed over twenty laks (2,000,000) of silver tangas.29 The limited nature of the

sultan's aims emerges from his dealings with Jajnagar (Orissa). Having mounted a brief attack on the kingdom of Shankara (Sarangarh), whose ruler fled,30 he advanced into Jajnagar, whose rai, Virabhanudeva

III, had ceased to send tribute.31 When Firuz Shah had uprooted the idol of Jagannath and obtained a

considerable booty, including a number of elephants, the Hindu king sent an offer of submission, and the

two monarchs performed a diplomatic minuet in which the rai claimed to have been the sultan's obedient

subject from the first and the sultan alleged that he had entered the country only for the purpose of hunting

elephants.

Firuz Shah's most successful campaign - though hardly a triumph32 -seems to have been that

against Nagarkot. The Hindu prince who had submitted to Muhammad b. Tughluq in 738/1337 had died,

and his son and successor repudiated the overlordship of Delhi. The sultan moved against him in c.

766/1365, and subjected the fortress to an investment of several months. At length the rai yielded and undertook to resume tribute payments. Firuz Shah treated the place with consideration, and notably

refrained from destroying the idol of Jawalamukhi at the rai's express request.33 The Nagarkot region

remained submissive thereafter, and would

27 cAfif, 254. Riazul Islam, 'Rise of the Sammas', 380. Inscription dated 784/1382 at Fath Masjid,

Paranrij, Sabarkantha district, in Desai, 'Khalji and Tughluq inscriptions from Gujarat', 21-2.

28 cAfif, 180.

29 TFS, 587-8. TMS, 124-5. cAfif, 111, mentions merely that the sultan conferred a chatr on the rai

of 'Chaparan': for the identification of this Hindu prince with Barani's rai of Gorakhpur, see Hodivala,

Studies, I, 311.

30 TMS, 129. Bihamadkhani, fols. 409b-410a (tr. Zaki, 8), describing Shankara as 'one of the great

cities of Jajnagar'. The city was identified by Hodivala, Studies, I, 387, and II, 149, on the basis of IM, 30.

Sarangarh lies 32 m. N.W. of Sambalpur, at 21° 36' N., 83° 7 E.

31 Thus according to the fath-nama reproduced in IM, 28. Both SFS, 54ff. (tr. Roy, 'Jajnagar

expedition', 62, 63-4), and cAfif, 163, comment on the prosperity of Jajnagar. For the identification of the

rai, see Hodivala, Studies, I, 318. The campaign is discussed in Banerjee, History of Firuz Shah, 40-2, and

in HN, 591-3.

32 R. C. Jauhri, 'A medieval invasion of Nagarkot (1363 A.D.)', JIH 44 (1966), 571-6.

33 SFS, 82-3. cAfif, 186-90. For the date, see Hodivala, Studies, I, 322.

serve as a valuable base for Firuz Shah's son Muhammad during the civil war of the early 1390s.

The acquisition of large numbers of elephants was clearly one of the desired objects of these

expeditions. Forty-seven were captured from Ilyas, forty were presented by Sikandar, and thirty-three were

taken in Jajnagar. According to Ibn Mahru, the rai of Jajnagar had ceased to send elephants to the sultan,

and the account of this campaign in the Sirat reflects the high priority given to the capture of elephants.34 In

addition, Bengal was agriculturally wealthy - a Chinese visitor in the 1340s testifies that reclamation had brought under the plough vast new tracts in the delta35 - and Ilyas had in 1346 made a lucrative raid on

Nepal.36 Firuz Shah also brought back large quantities of silver, to which, as we saw (p. 261), Bengal had

access in plenty and which was still in short supply within his own dominions.37 Perhaps the sultan, who

allegedly observed to Tatar Khan in 754/1353 that his predecessors had reduced Bengal but had proved

unable to control it in view of the nature of the terrain,38 aimed no higher than replenishing his treasury and

the pllkhana.

The great majority of the operations against the Hindus of which cAfif speaks did not involve the

sultan's participation, being presumably conducted under the aegis of the local muqtacs; and Sirhindi

praises the great amirs in the direction of 'Hindustan' for chastizing 'rebellious infidels' and maintaining the

authority of Delhi.39 But conditions may have deteriorated during the sultan's last twelve years, when he is

supposed to have abandoned military activity but we suddenly see him campaigning personally in Katehr and Etawa for the purpose of securing tribute; cAfif, taking his cue from Firuz Shah himself, dresses up

these attacks as hunting expeditions.40 In 787/1385-6 the sultan constructed the new fortress of Firuzpur at

Biuli (Beoli), some fifteen miles from Bada'un, as part of his defence measures in the region.41 In Etawa,

during a campaign in 779/ 1377-8, the two muqaddams, Sumer and Uddharan, were taken to Delhi, and

fortresses were built at Akhal (renamed Tughluqpur) and *Patlahi. A new fortress at Firuzpur (near Kanar)

became the centre of a new shiqq, incorporating Tughluqpur and Rapri, which was entrusted to Malikzada

34 IM, 28. SFS, 54, 58, 63 (tr. Roy, 'Jajnagar expedition', 61-2, 65). cAfif, 123, 161, 163, 167. 171;

cf. also 172, 175, for the total of seventy-three elephants from Bengal and Jajnagar. IM. 32, gives fifty-three

as the number of elephants surrendered by the rai of Jajnagar. TFS, 592, 594, gives the total number of

elephants taken on the first Bengal expedition as forty-four. 35 W. W. Rockhill, 'Notes on the relations and trade of China with the eastern archipelago and the

coast of the Indian Ocean during the fourteenth century', T'oung Pao 16 (1915), 435-6. citing the Tao-i

chih-lueh of Wang Ta-yiian (1350).

36 K. P. Jayaswal, 'An unrecorded Muhammadan inscription of Nepal', JBORS 22 (1936). part 2,

93-5. Dani, Muslim inscriptions of Bengal, 129.

37

TFS, 597. Simon Digby, 'The Broach coin-hoard as evidence of the import of valuta across the

Arabian Sea during the 13th and 14th centuries', JRAS (1980), 129, 135-6 n.4. 38 cAfif, 119.

39 TMS, 133.

40 cAfif, 493, 497.

41 TMS, 135. For the location of Beoli, see Hodivala, Studies, I, 389.

Firuz (later wazir to Sultan Tughluq Shah II), son of Taj al-Din Turk: it was to become the nucleus

of the principality of Kalpi.42

The sultan and the nobility

Both cAfif and Sirhindi point to the fact that Firuz Shah's era witnessed only one revolt (namely,

by a Muslim noble), that of Shams al-Din Damghani in 782/1380-1, and the former expressly draws a

contrast with the turbulence of Muhammad b. Tughluq's reign.43 Damghanl, appointed na'ib of the iqtac of

Gujarat in 778/1377, failed to realize the enormous sums for which he had contracted, omitted to despatch

to court any of the revenue he had raised, and rose in open rebellion. His extortionate measures, however,

had alienated those on whose support his revolt depended, the amiran-i sada of Gujarat, who remained

loyal to the sultan; they attacked and killed him. Apart from this episode, which cAfif dismisses as a farce,44

the absence of revolt by Muslim nobles bears out the picture of the reign furnished by our sources, as an era

of contentment among the aristocracy.45

Yet such remarkable quiescence was obtained at a price. Unlike his predecessor, Firuz Shah was

notoriously uninterested in the details of day-to-day fiscal administration,46 and almost from the beginning

left the conduct of affairs to the wazir Khan Jahan. The new wazir proved to be the main pillar of the

regime. During Firuz Shah's absences on campaign, he contrived to overawe the capital with

demonstrations of military force, and skilfully concealed from the citizens the lack of news from the sultan

when the Delhi army got lost on the way back from Jajnagar and again when Firuz Shah found himself in

the Rann of Kachh.47 But as Barani had pointedly observed, Khan Jahan enjoyed more extensive powers

than had been vouchsafed to any previous wazir, and cAfif quotes the sultan himself as saying that Khan

Jahan was the real ruler of Delhi.48 When the wazir clashed with cAyn al-Mulk Ibn Mahru, who was then

serving as accountant-general (mushrif-i mamalik), he got his own way by threatening to leave for Mecca.

The sultan capitulated and gave Khan Jahan permission to employ and dismiss whomsoever he wished. Ibn Mahru thus forfeited his post; though when shortly granted the iqta's of Multan, Siwistan and Bhakkar, he

secured from Firuz Shah the concession that the finances of these territories would lie outside the wazir's

jurisdiction.49

Firuz Shah seems to have assigned a significantly higher proportion of

42 Bihamadkhani, fol. 412b (tr. Zaki, 13-14). TMS, 133-4.

43 cAfif, 492-3; 497 for the date of the revolt, which is discussed in Hodivala, Studies, I, 388-9.

TMS, 132, gives 778, the year of Damghani's appointment.

44 cAfif, 499-502.

45 Ibid., 288, 297-8.

46 Ibid., 341-2.

47 Ibid., 173, 211-14, 398-9.

48

TFS, 578-9. cAfif, 400; cf. also 411.

49 Ibid., 408-14.

the Sultanate's territories as iqtac, thus reducing the extent of the khalisa. The nobility as a whole,

moreover, now enjoyed greater privileges than in the time of Muhammad b. Tughluq. We may perhaps

explain the fact that the salaries of khans and maliks were now considerably higher than in the previous

reign as a measure designed to protect them against inflation (see below, p. 316).50 But already, within the

first six years of Firuz Shah's era, Barani could observe that the revenue from the iqta's was not audited

with the same rigour as before; and cAfif tells us specifically that the sultan departed from previous practice

in having the annual gifts presented by the provincial governors valued and offset against the revenue-

demand from their territories. Certain of the amirs, notably the 'arid cImad al-Mulk Bashir, accumulated

enormous fortunes.51

It was the sultan's policy also to allow the heir of an amir, a muqtac or an official to inherit his

father's position, title, and iqta's or other emoluments. In some measure this pattern of inheritance had

obtained previously -certainly in the thirteenth century (pp. 101-2 above), even if it had perhaps been

attenuated under cAla' al-Din and discontinued by the first two Tughluqids. But under Firuz Shah it

undoubtedly became the norm. When Khan Jahan (I) died, for instance, his office of wazir and his title both

passed to his son Jawnan, who was henceforth known as Khan Jahan (II); Zafar Khan, the refugee amir

from Bengal, was succeeded in his iqtac of Gujarat by his son Darya Khan, who was likewise given the title

of Zafar Khan (II); and the rank and title of the "arid cImad al-Mulk Bashir passed on his death to his son

Ishaq.52 Such examples could be multiplied, and they extended to all levels of the bureaucracy.53 Even in an

emergency, as when he transferred Malik Nasir al-Mulk Mardan Dawlat from the east to the Multan

frontier to deal with the Mongol threat, Firuz Shah did not disregard the hereditary principle: Nasir al-Mulk's iqta's of Kara and Mahoba were simply assigned to his adopted son Sulayman.54 Nasir al-Mulk's son

Malik Shaykh followed him for a short time in the command at Multan; and after his death, Sulayman was

transferred to Multan, where he in turn was soon succeeded by his own son Khidr Khan.55

The allusion to the system of hereditary offices in the Futuhat-i Firuz-Shahi indicates that the

sultan took some pride in it, and it is laid to his credit also by the author of the Sirat.56 But that such a

policy was likely to

50 Ibid., 296-7'. Habib, 'Agrarian economy', 73.

51 TFS, 555-6. cAfif, 268-9. For Bashir's wealth, see ibid., 438, 439-40, 445.

52 Khan Jahan: ibid., 425-6. Zafar Khan: ibid., 286, 499; TMS, 131. cImad al-Mulk: cAfif, 445.

SFS, 153-4, gives a list of such hereditary appointments, headed by Khan Jahan (I).

53 For two instances from the diwan-i wizarat, see cAfif, 482.

54 TMS, 133.

55 Ibid., 182. HN, 632, may be wrong in emphasizing that the appointments of Mardan's

successors dated from after Firuz Shah's death; Khidr Khan, who received his title in 791/ 1389, had

previously, as muqtac of Multan, himself borne the style of Nasir al-Mulk: TMS. 146, 147.

56 FFS, 18 (tr. Roy, 463). SFS, 153-5.

implant in the diwan-i wizarat, for example, officials who were unequal to their task was the

complaint of Shams al-Din Abu-Rija when auditor-general (mustawfi-yi mamalik).51 When applied to

provincial governorships, it would in time have the effect of creating entrenched regional interests and the

autonomous principalities which emerged in the era of Firuz Shah's grandsons.

The first civil war

Of the events of the crisis that began in the twilight years of Firuz Shah's reign, we learn a good deal from two authors, Sirhindi and Bihamadkhani.58 But the wherewithal to explain it is more elusive; and

here cAfif, through occasional references to the troubles in his Ta'rikh-i Firuz-Shahi, provides greater

insight. It seems that the activities of Shams al-Din Abu Rija, who had made enemies of all the amirs and

whose disgrace in 786/1384 preceded the crisis, had already thrown the state into confusion,59 perhaps

because the nobility saw its fiscal autonomy as under threat. It needs also to be borne in mind that Firuz

Shah was by now relatively advanced in age (he was eighty-three lunar years old when he died in

790/1388) and had been ill since 786/1384, and according to cAfif most or all of his contemporaries among

the grandees had predeceased him.60 The links that bound together the sultan and the new generation of

amirs had presumably slackened. But what especially brought disaster on the empire was, in cAfif's view,

the rivalry between the wazir, Khan Jahan (II), and Firuz Shah's son Muhammad.61

During the sultan's last years, the wazir was able to exercise virtually untrammelled power and used the opportunity to remove various amirs and maliks who opposed him.62 The death of Firuz Shah's

grandson Fath Khan in 778/1376 had been a heavy blow to him,63 and Bihamadkhani claims that he

selected as his heir his great-grandson, Fath Khan's son Tughluq Shah (see appendix VI).64 When in

789/1387 the wazir endeavoured to turn the subservient sultan against his only surviving son, Prince

Muhammad, a crisis arose; he was obliged to flee from Firuzabad and was subsequently killed in Meo

territory. Those of the wazir's adherents who were executed after his . overthrow included Malik Bihzad-i

Fathkhani, presumably a former slave of Prince Fath Khan,65 and it may be that we are dealing

57 cAfif, 474-5.

58 For what follows, see generally HN, 618-22; Husain, Tughluq dynasty, 441-50. 59 cAfif, 455-6, 457, 459, 492, 498.

60 Ibid., 444-5, 497 (where the majority are said to have died in 781/1379-80), 498. For a list of

leading nobles who died before 772/1370-1, including Khudawand Khan, Da'ud Khan b. Bayyu, Ibn Mahru

and Ikhtiyar al-Din-i Nuwa, see SFS, 154. For the year of Firuz Shah's birth, see above, p. 249, n.70.

61 cAfif,427.

62 TMS, 135-6.

63 cAfif, 494. 64 Bihamadkhani, fol. 414a (tr. Zaki, 16).

65 TMS, 137.

simply with two groups that had coalesced around Muhammad and around the descendants of his

eldest brother. Muhammad was now made wazir and then enthroned as joint sultan in Sha'ban 789/August

1387. At this stage, we are told, he enjoyed the sympathy of not only the amirs and the people of the capital

but also Firuz Shah's slaves.66 This group may have joined him out of hatred for the wazir, since the

principal slaves are said to have been alarmed at an earlier date by the growing power of Khan Jahan (I).67

But after five months the slaves turned against Muhammad, according to Sirhindi, out of antipathy towards his favourites, Sama' al-Din (now entitled Mucin al-Mulk) and Kamal al-Din (Dastur Khan), the two sons of

Malik cUmar, the 'arid-i bandagdn-i khass; although cAfif, obscurely, attributes the change of allegiance to

the enormous sums left by the old sultan's 'arid, Malik Bashir cImad al-Mulk.68

Muhammad in turn was expelled from the capital and retired to Nagarkot, and the old sultan now

recognized as co-ruler Tughluq Shah, who was enthroned as Sultan Ghiyath al-Din when Firuz Shah died

on 18 Ramadan 790/20 September 1388. On 21 Safar 791/20 February 1389, however, Tughluq Shah and

his wazir, Malikzada Firuz (Firuz Khan) b. Taj al-Din Turk (ancestor of the later rulers of Kalpi), were

killed in a rising by the na'ib-wazir Rukn al-Din Junda (p. 190 above), who put Abu Bakr Shah, a grandson

of Firuz Shah, on the throne. Junda was promoted to wazir, but shortly plotted to remove Abu Bakr in turn, perhaps proclaiming another of Firuz Shah's grandsons, Firuz Shah b. Zafar,69 and was killed. Over the next

two years Muhammad, based first at Samana and then at Jalesar in the Doab, sent troops to ravage the

territory surrounding Delhi and himself made three unsuccessful attempts on the capital.

The situation, as Sirhindi observed, was one of stalemate, for Abu Bakr could not be dislodged but

when victorious was unable to leave the capital and pursue his enemy; with Muhammad, on the other hand,

were ranged 'all the amirs, maliks, troops (hasham), retainers (khadam) and subjects (ra'aya) of the

empire'.70 Abu Bakr's strength lay in his possession of the capital and the elephantry and in the allegiance of

his grandfather's slaves.71 Muhammad, who recognized them as the principal obstacle to his success,

ordered his adherents in the provinces to arrest and kill all the old sultan's slaves on 19 Ramadan 791/11

September 1389.72 This mass execution testifies to the widespread support for Muhammad among the

military class outside the Delhi complex. Tughluq Shah had exiled Ghalib Khan, the son 66 Ibid., 136-7.

67cAfif,415.

68 TMS, 139. cAfif, 440.

69 'Note on a gold coin bearing the name of Prince Firuz Shah Zafar, son of Firuz Shah of Dihh",

JASB 40 (1871), 160. Thomas, 300. But both authors assume incorrectly that the prince named is the old

sultan's son Zafar and that this issue of 791 was therefore posthumous. Cf. also CMSD, 191-4 (nos. 771-

779A), 223-4.

70 TMS, 148.

71 Ibid., 146. Bihamadkhani, fols. 421b, 422a (tr. Zaki, 30, 31).

72 TMS, 146; for the massacre, see 147.

and successor of Malik Qabul 'Qur'an-khwan', and deprived him of Samana; but his own nominee

as muqta' was killed in Safar 791/February 1389 by the amiran-i sada of that territory, who invited in

Muhammad, and Ghalib Khan was restored.73 The commanders who joined Muhammad with their forces

for his various attacks on Delhi included the muqta's of Multan and Bihar, the sons of the governors of

Qinnawj and Awadh, the Ma'in and Bhatti chiefs who held iqta's in the eastern Panjab, muqaddams from the hills (presumably the Qarachil foothills) and rais and ranas from Etawa.74 Support for Abu Bakr is

found only in Alwar, where he could count on the Meo chieftain Bahadur Nahir.75 In Gujarat the na'ib, the

Firuz Shahl slave Malik Mufarrij Sultani, had in 789/1387 killed Sikandar Khan, newly arrived as governor

on Muhammad's behalf, and had been recognized by Tughluq Shah as governor with the style of Rasti

Khan; but whether he transferred his allegiance to Abu Bakr is unknown.76

It is unclear why in Ramadan 792/August 1390 a split emerged within the slaves' ranks, and a

group of them, headed by Islam Khan Mubashshir-i Chap Sultani, invited in Muhammad. The fact that the

prince's own party included Firuz Shahi slaves - Malik Sarwar Sultani, the shihna of Delhi, whom

Muhammad had made wazir with the style of Khwaja Jahan, appears throughout as his loyal adherent77 -

may help to explain the readiness of certain of the slaves in Abu Bakr's camp to give him possession of the capital. Bihamadkhani, moreover, says that Malik Shahin Sultani, entitled cImad al-Mulk, the former

commander of Firtiz Shah's pilkhana, had been driven from the city by the amirs responsible for the

overtures to Muhammad,78 suggesting that the change of allegiance may have been connected with rivalry

between Islam Khan and cImad al-Mulk. Once securely in possession of the elephantry, however, and

possibly after offering a guarantee to his new ally Islam Khan, Muhammad had all the slaves expelled from

Delhi or put to death.79 In 794/1392 Khwaja Jahan Sarwar, whom Islam Khan had replaced as wazir,

trumped up charges against his rival; and despite his services in the final engagement with Abu Bakr, Islam

Khan was executed.80 We still hear occasionally of Firuz Shahl

73 Ibid., 140, 145; for Ghalib Khan restored in Samana, ibid., 147, 156. His genealogy is given in

Bihamadkhani, fol. 431b (tr. Zaki, 45).

74 TMS, 145, 146-7.

75 Ibid., 146, 149, 151; he had supported Tughluq Shah also (ibid., 142). Bihamadkhani, fols.

423a, 426 (tr. Zaki, 32, 35).

76 TMS, 138; also 142 for Tughluq Shah's recognition.

77 Ibid., 146, 147, 152-3. Bihamadkhani, fol. 421b (tr. Zaki, 30). cAfif, 338, says that he was in

charge of Firuz Shah's jewel-house (jawdhir-khana).

78 TMS, 148, 149-50. Bihamadkhani, fols. 421a, 423a (tr. Zaki, 29, 32). For Shahln, see also cAfif,

338.

79 TMS, 149-50. Bihamadkhani, fol. 425b, saying that the sultan cahd-u payman ba-tajdid dar

award with Islam Khan (tr. Zaki, 34).

80 TMS, 152-3. Bihamadkhani, fol. 430b (tr. Zaki, 44), gives a briefer account of the plot, but does

not mention Khwaja Jahan's role.

slaves thereafter;81 but Muhammad's accession marks the destruction of this highly volatile

element as a force capable of making and unmaking sultans.

According to Bihamadkhani, Muhammad's triumph brought about peace and repose.82 Abu Bakr,

expelled from Delhi, fled to Alwar, where his uncle's forces defeated him in Muharram of the following

year/December 1390; he shortly died as a prisoner at Amroha. The sultan also acted quickly to restore his

authority in Gujarat, where in Safar 794/January 1392 the hostile Mufarrij (Rasti Khan) was defeated and

killed by Muhammad's appointee, Zafar Khan Wajih al-Mulk.83 But Bihamadkhani's encomium can only

have applied to the capital and the more westerly provinces, since the sultan had to spend the rest of his

reign fighting Hindu chieftains in Alwar and the Doab.

Loss of territory to the infidel

'Thanks to the contest among the Muslims for the sovereignty,' says Sirhindi, 'the infidels of Hindustan gained in strength, refrained from paying the jizya and the kharaj and plundered the Muslim

townships (qasabat).84 As we have seen, the old sultan's last years had not been free of disturbances, but

Hindu princes appear to have asserted themselves more vigorously in the wake of Muhammad's expulsion

from Delhi in 789/ 1387. There are signs of a struggle with the local Hindus around Nagawr in that year

and again in 791/1389, when the na'ib of the shiqq of Nagawr and Jalor was killed.85 We are more fully

informed about conditions in the southern Doab, as in Etawa for instance, and the territories south of the

Yamuna which would later form the independent principality of Kalpi. In Etawa, Uddharan and Sumer,

who had supported Muhammad, returned home and rose in revolt following Tughluq Shah's accession,

inflicting a heavy defeat on Malik Mahmud, who now governed the shiqq of Firuzpur in succession to his

father, the wazir Firuz Khan b. Taj al-Din Turk. Tughluqpur was surrendered to the enemy, and the towns

of Chandawar, Bhonga'un and Rewa, among others, all fell into the hands of Hindu princes. For the moment, Malik Mahmud was able to do no more than occupy Kalpi, which he renamed Muhammadabad,

in 792/1390 and to make it his headquarters.86

In 794/1391-2 Sultan Muhammad took the field against the enemy in

81 TMS, 160. Yazdi, ZN, ed. Ilahdad, II, 64/ed. Urunbaev, fol. 310a, however, seems to apply the

term (wrongly) to Mallu Iqbal Khan and his associates.

82

Bihamadkhani, fol. 424b (tr. Zaki, 33).

83 Harawi, III, 83, 84-5. Misra, 141-2, citing the Tabaqat-i Mahmud-Shahi and the Mir'at-i

Sikandarl, respectively fifteenth- and sixteenth-century chronicles of Gujarat.

84 TMS, 147.

85 ARIE (1975-6), 163 (D188). ARIE (1969-70), 93 (D167).

86 These events are described only by Bihamadkhani, fols. 418b-419b (reaDing RYWH for RTWH),

436b-437a (tr. Zaki, 26-7, 54).

Etawa, where Uddharan and Sumer had sacked Balaram. Having razed Etawa to the ground, the

sultan moved back across the Ganges and chastised the Hindus of Qinnawj and Dalmaw, building a fortress at Jalesar, which he renamed Muhammadabad. In Alwar Bahadur Nahir, who had consistently sided with

Muhammad Shah's enemies, continued to defy him and had to be driven from Kotla; in the west, the

Khokhar chief Shaikha rebelled and sacked Lahore in 796/1394, and a punitive expedition was in

preparation under Muhammad Shah's son Humayun Khan when the sultan died.87

The second civil war

Muhammad did not long enjoy the throne for which he had mounted such a determined struggle,

dying on 17 Rabi I 796/20 January 1394.88 Humayun Khan, who succeeded him as cAla' al-Din Sikandar

Shah, followed him to the grave on 5 Jumada 1/8 March, and a younger son was thereupon proclaimed

sultan as Nasir al-Din Mahmud Shah. The new monarch was to prove little more than a cipher. There is a suggestion that his enthronement commanded scant support, for the wazir Khwaja Jahan Sarwar had to

persuade the amirs whose territories lay to the west, like Ghalib Khan of Samana and Rai Kamal al-Din

Ma'in, not to leave Delhi without pledging their allegiance.89 But in Rajab 796/May 1394 Khwaja Jahan

was sent east with an army, twenty elephants and the title of Sultan al-Sharq, and entrusted with the

territories 'from Qinnawj to Bihar', so that he might clear the region of recalcitrant Hindu chieftains.90 He

set up his headquarters at Jawnpur and never returned to Delhi. Power at court was disputed among a

number of war-lords, notably Muhammad (Tatar Khan), son of Wajih al-Mulk Zafar Khan the governor of

Gujarat, and a group whom Bihamad-khani calls Muhammad Shah's more important slaves (bandagan-i

kibar), notably Muqarrab al-Mulk (styled Muqarrab Khan), cAbd al-Rashid Sultani (entitled Sa'adat Khan)

and Mallu (later Iqbal Khan). Sa'adat Khan was ousted by a conspiracy in which Mallu was implicated, and

took refuge with Tatar Khan. Having lost possession of the sultan, Tatar Khan's party in Muharram

797/October 1394 enthroned at Firuzabad a younger brother of Tughluq Shah II as Nasir al-Din Nusrat Shah.91 There were now once more two rival sultans, each with his own capital

87 TMS, 154.

88 For the second civil war, see generally HN, 623-5; Husain, Tughluq dynasty, 452-60; Lal,

Twilight, 8-12.

89 TMS, 156.

90 Ibid., 156-7. His career is traced in Mian Muhammad Saeed, The Sharqi Sultanate of Jaunpur

(Karachi, 1972), 20-35. 91 TMS, 158-9. Bihamadkhani, fols. 432b-433a (tr. Zaki, 47-8), alone refers to Mallu and the

others as slaves of Muhammad Shah; he does not mention the conspiracy against Sacadat Khan, and gives

greater prominence to Tatar Khan's role than does Sirhindi. Nusrat Shah's

city and military establishment and each a puppet in the hands of powerful grandees. This

situation persisted for three years, with fighting between the two sides an almost daily occurrence.

According to Sirhindi, Mahmud Shah's party controlled only Old Delhi and Siri. At Firuzabad, Nusrat Shah

and Tatar Khan commanded the allegiance of the Doab, together with Sonpat, Panipat, Jhajhar and

Rohtak.92 The omission of any of the Sultanate's other territories demonstrates how little impact events at the centre now had on the governors of major provinces, although coins and inscriptions continued to

indicate a nominal allegiance: thus Mahmud Shah was recognized in the regions controlled by Khwaja

Jahan, whereas Zafar Khan in Gujarat acknowledged Nusrat Shah.93

When recounting Temur's invasion a few years later, the Timurid chronicler Sharaf al-Din Yazdi

was under the impression that Mallu and his brother Sarang Khan, since Mahmud Shah's accession

governor of Deopalpur, were the real masters of his empire;94 and Professor Hambly has shown how the

partnership sought to dominate the Sultanate from two distinct bases, Sarang Khan in the Panjab and Mallu

Iqbal Khan at Delhi.95 From Deopalpur Sarang Khan had embarked in 798/1395-6 on a sustained effort to

bring the neighbouring territories under his own control (and hence, very indirectly, under that of Mahmud

Shah). Shaikha was defeated, and Lahore reoccupied.96 For a time Sarang Khan also held Multan; the

muqtac, Khidr Khan, was taken prisoner but later escaped. But when Sarang Khan attacked Samana, Ghalib Khan appealed to Tatar Khan, who in Muharram 800/October 1397 defeated Sarang Khan and drove him

back to Multan, reinstating his protege in Samana.97 In Dhu'l-Qacda/October-November 1398 Mallu, who

had meanwhile briefly declared for Nusrat Shah only to seize control of his elephants and had then put

himself at the head of the rival group by the elimination of Muqarrab Khan, moved against Tatar Khan's

base at Panipat, which he captured. Tatar Khan, weakened by the desertion of prominent supporters, retired

to his father in Gujarat.98 Mallu Iqbal Khan was thus left in undisputed control of both

laqab appears as Shams al-Din in an inscription of 797/1395: Desai, 'Khalji and Tughluq

inscriptions from Gujarat', 37-8.

92 TMS, 159-61. Bihamadkhani, fol. 433 (tr. Zaki, 48). 93 Desai, 'Khalji and Tughluq inscriptions from Gujarat', 34-8. Sayyid Yusuf Kamal Bukhari,

'Inscriptions from Maner', EIAPS (1951-2), 15-16.

94 Yazdi, ZN, ed. Ilahdad, II, 14-15/ed. Urunbaev, fol. 296a (duplicated at 301a). For Sarang

Khan's appointment, see TMS, 156.

95 Gavin R. G. Hambly, 'Twilight of Tughluqid Delhi', in Frykenberg (ed.), Delhi through the

ages, 47-56.

96 TMS, 157-8.

97 Ibid., 161-2. Khidr Khan's capture is mentioned only by Yazdl, ZN, ed. Ilahdad, II, 175 (omitted

in Urunbaev edn, fol. 341a).

98 TMS, 163-5. Bihamadkhani, fols. 433b-434b (tr. Zaki, 49-50). Fadl-Allah Balkhi as Mallu's

lieutenant in 801/1398: Ghiyath al-Din Yazdl, tr. Semenov, 121; Shami, ZN, I, 191: Yazdi, ZN, ed. Ilahdad,

II, 116/ed. Urunbaev, fol. 324b. TMS, 160, earlier lists him among Nusrat Shah's adherents.

capitals, Delhi and Firuzabad, and of Sultan Mahmud Shah. The chronology of his various acts of

duplicity suggests that he was attempting to shore up his position in reaction to the elimination of his

brother Sarang Khan at Multan by Temur's forces (p. 313 below). Only a few weeks later, however, Mallu

in turn was effectively swept away by Temur."

The north-west frontier and Temur's invasion

The sources depict Firuz Shah's reign as relatively free of Mongol attacks.100

Nevertheless, they

appear still to have been a regular occurrence. Barani mentions just two minor inroads in the period of six

years before he ceased writing.101 One took place in the neighbourhood of the Sodra river (the Chenab),

while the other, into Gujarat, which was checked partly by the sultan's troops and in part by the muqaddams

of the region, may have been connected with the encouragement given to the Mongols by the Samma

prince Banbhina, about which Ibn Mahru complains.102

Sirhindi tells us briefly that towards the end of

759/1358 the Mongols invaded the Deopalpur territory but withdrew on the advance of the sultan's forces under Malik Qabul ('Qur'an-khwan').103 The Sirat, lastly, claims that the Mongols were in the habit of

advancing to the Beah and harassing the villages, but refers to a defeat inflicted on them by the army of

Delhi in the year of Firuz Shah's Nagarkot campaign (i.e. c. 767/1365-6).104 The sultan was sufficiently

anxious about the Mongol frontier to transfer there from the east Nasir al-Mulk Malik Mardan Dawlat,

because he allegedly had no one else of the calibre necessary to deal with the Mongol danger.105

More than this we are not told; nor are the attacks we know of easily linked up with events in the

Chaghadayid territories. Here the death in 759/ 1358 of Muhammad b. Tughluq's ally, the Qara'unas noyan

Qazaghan, had inaugurated a lengthy period of strife among the clan leaders and provoked two brief

interventions in Transoxiana by the eastern Chaghadayid khan, Tughluq Temur. Temur, a member of the

Turco-Mongol clan of the Barlas, collaborated for a time against the invaders with Qazaghan's grandson

Husayn; but the allies shortly fell out, and in 771/1369-70 Temiir vanquished Husayn and replaced him as the real ruler of the western Chaghadayid ulus.106 The effects of these upheavals were felt in the Indian

99 Hambly, 'Twilight', 50-1, similarly proposes that the activities of Mallu and Sarang Khan were

closely connected. But I am not convinced by his suggestion that Sarang Khan's move against Samana was

part of a plan to join forces with his brother in order better to resist the imminent invasion of Pir

Muhammad. Sirhindi's chronology makes it clear that Sarang Khan was eliminated before Mallu embarked

upon his complicated intrigues between the two rival sultans. 100 cAfif, 321.

101 TFS, 601.

102 IM, 101; and see also 230. 103 TMS, 127. Qur'an-khwan subsequently became muqtac of Samana (p. 187 above).

104 SFS, 285-6.

105 TMS, 133: the context suggests the 1370s.

106 Beatrice Forbes Manz, 'The ulus Chaghatay before and after Temur's rise to power: the

borderlands as well as in Transoxiana. In 763/1361-2 Tughluq Temur's army is said to have

plundered the territory as far as the Hindu Kush.107 Qazaghan's sons had fled to Kabul and Ghazna on their

father's murder, and the region seems to have served as the power-base of Husayn, who was active there in 761/1360 and later, with Temur's aid, recovered Kabul from his enemies.108 We have seen (pp. 224, 228-9)

how Mongol amirs sought refuge across the Indus during the early years of the century; and commanders

who lost out in these fresh conflicts likewise turned towards India, as Husayn at one point contemplated

doing and as his sons did following his overthrow in 771/1369-70.109 But our Indian sources supply too

little detail to enable us to make any connections with the few Mongol inroads they record.

Once Temur had supplanted Husayn as de facto ruler of the western half of Chaghadai's ulus, it

was vital for him to absorb the energies of the tribes in external campaigns; and this was also a means of

denying a refuge outside the ulus to dissident noyans.110 But Temur, who was not of Chinggisid blood and

who ruled the ulus through a puppet khan (actually chosen from the line of Ogodei), appears to have seen it

as his task to reconstitute Chinggis Khan's empire, though largely in the form of protectorates under Chaghadayid overlordship.111 With this in view, he launched attacks on the Kartid kingdom of Herat,

whose 'Tajik' ruler had displayed the effrontery to assume the style of sultan (p. 235 above); on the various

other powers that had sprung up amid the ruins of the Ilkhanate; and on the Golden Horde.112 Although the

justification given for his invasion of India towards the end of the century was religious and couched in

terms of the spread of Islam, it can only have been a facade: the most that can be said on this count is that

the aim was perhaps to punish Muslim rulers for permitting such licence to their vast numbers of Hindu

subjects and servitors.113

According to Bihamadkham, Temur and Firuz Shah had corresponded, and it may be for this

reason that Muhammad b. Firuz at one point thought

transformation from tribal confederation to army of conquest', CAJ21 (1983), 86-95, and her Rise

and rule, 41-57.

107 Shami, ZN, I, 18-19. Yazdi, ZN, ed. Ilahdad, I, 59/ed. Urunbaev, fol. 100a. P. Jackson, Tughluk

Temiir', Enc.Isl2.

108 Natanzi, 197. Shami, ZN, I, 51. Yazdi, ZN, ed. Ilahdad, I, 48, 175/ed. Urunbaev, fols. 97a,

130b.

109 Shami, ZN, I, 31. Yazdi, ZN, ed. Ilahdad, I, 71, 206/ed. Urunbaev, fols. 103a, 139a.

110 Manz,'Ulus Chaghatay', 98.

111 Hans Robert Roemer, 'Timur in Iran', in P. Jackson and L. Lockhart (eds.), The Cambridge

history of Iran, VI. The Timurid and Safavid periods (Cambridge, 1986), 52, 57, 72.

112 For these campaigns, see ibid., 46-73; Tilman Nagel, Timur der Eroberer und die islamische

Welt des spaten Mittelalters (Munich, 1993), 377-86; a brief survey in Manz, Rise and rule, 67-73.

113 See, e.g., Ghiyath al-Din Yazdi, Ruz-Nama, tr. Semenov, 60; ShamI, ZN, I, 170; Yazdi, ZN, ed.

Ilahdad, II, 15/ed. Urunbaev, fol. 296a.

of abandoning the struggle against Abu Bakr Shah and seeking Temur's assistance; he had actually

set out for Samarqand with a small group of followers when he was invited to come to Delhi and take the

throne.114 Although the journey to Transoxiana proved unnecessary, it is possible that some message had

been despatched to Samarqand in advance. But in all likelihood Temur needed no invitation to intervene in

the chaos within the Delhi Sultanate, which presented him with an ideal opportunity for plunder.115

Temur's grandson Pir Muhammad, who governed much of present-day Afghanistan from Kabul,

crossed the Indus in Rabi I 800/November-December 1397 and defeated the troops sent to relieve Uchch by

Sarang Khan, who was then himself forced to surrender Multan in Ramadan/June 1398. Pir Muhammad

established his headquarters in the city.116 Temur arrived in the Multan region in mid Safar 801/late in

October. Sending his main force by way of Deopalpur and Samana, he marched via Bhatner and Sarsati,

putting both strongholds to the sack, before rejoining the rest of his troops on the banks of the Ghaggar. On 7 Rabi II 801/16 December 1398 he did battle with Mallu Iqbal Khan and Mahmud Shah in the plain

outside the capital. Although the Indian army put up a brave fight, it was routed. The sultan and Mallu

withdrew into the city, and shortly fled, Mallu into the Doab and Mahmud Shah to Gujarat, while the

khutba in Delhi was read in the name of Temur's nominal sovereign, the Ogodeyid Mahmud Khan.117 The

amnesty granted to the citizens of Delhi meant nothing once Temur's troops were inside the city and

disorders broke out: the sack began on 9 Rabi 11/18 December and lasted for some days. After

campaigning east of the Yamuna, where he stormed Mirat (which had successfully withstood Tarmashirin

seventy years previously) and launched an unsuccessful attack on the fortress of Hardwar, Temur finally

withdrew westwards through the foothills, attacking Jammu en route (middle of Jumada II/late February

1399).118 For all his posturings, his invasion had enveloped Muslim amir and Hindu chief alike in a

common destruction. According to Bihamadkhani, Sarang Khan had been put to death; Bahadur Nahir, who had submitted to him after the sack of Delhi, may have been put in chains, and the Khokhar chief Shaikha,

who had acted as guide to the invaders, was arrested with his family during the conqueror's return march.119

It may well be asked what enabled Temur to succeed - to defeat the army of the Sultanate and to

capture its capital - where his Chaghadayid

114 Bihamadkhani, fols. 422b-423a (tr. Zaki, 32); for Firuz Shah and Temur, see fol. 442b (tr. Zaki,

59-60).

115 The view of Roemer, 'Timur in Iran', 70. 116 TMS, 162-3.

117 Shami, ZN, I, 192; and see Woods, 'Rise of Timurid historiography', 104-5.

118 For a detailed survey of the Indian campaign, see Lal, Twilight, 16-40.

119 Yazdi, ZN, ed. Ilahdad, II, 127-8/ed. Urunbaev, fols. 327b-328a. TMS, 166-7. Bihamadkhani,

fol. 307a (tr. Zaki, 93).

predecessors had failed. The reasons are manifold. One was that the conqueror had at his disposal

resources of revenue and manpower that had not been available to Chaghadayid princes like Qutlugh Qocha and Tarmashirin, since his campaigns in Persia and against the Golden Horde had won for him

tribute and contingents of troops from areas that had lain outside Chaghadai's ulus earlier in the century; he

had also welded the Chaghadayid tribal forces into a far more formidable war-machine.120 But it is still

more important to register the sharp decline that had occurred in the military establishment of the Delhi

Sultanate under Firuz Shah and his successors.

The decline in the Sultanate's resources

Timurid authors - by no means inclined, we can be sure, to minimize the opposition that their hero

vanquished outside Delhi - set the army with which Mahmud Shah and Mallu met him at 10,000 horsemen,

20,000 foot and 120 elephants.121 These numbers constitute a pitiful force compared with those that had accompanied Firuz Shah on campaign. For his two invasions of Bengal, that sultan had been able to raise

armies of 80,000 or 90,000 horsemen and 450 elephants;122 for his Thatta expedition, 90,000 horse and 480

elephants.123 cAfif, probably indebted for these figures to his father, who worked in the diwan-i wizarat,

tells us at another juncture that the sultan possessed a total of 80,000 horsemen excluding his slaves.124 Yet

even such statistics as these are a pale reflection of the numbers on the muster-roll under 'Ala' al-Din Khalji

or in the early years of Muhammad b. Tughluq.

Blame for the unimpressive military establishment by the time of Temur's invasion cannot all be

heaped upon Firuz Shah. The drop in the number of elephants in all probability reflects the fact that those

animals formerly despatched to the Delhi Sultan as tribute from Bengal and Jajnagar were now being sent

instead to Khwaja Jahan at Jawnpur.125 Even where decline can be traced to his era, it would be foolish to

disregard circumstances over which Firuz Shah had no control. Security from external attack brings its own penalty, as cAfif recognized, in a deterioration in the quality of the military.126 In this process the decline in

the incidence of Mongol attacks

120 Manz, Rise and rule, chapters 4-5.

121 Ghiyath al-Din Yazdi, Ruz-Nama, tr. Semenov, 115. Shami, ZN, I, 189. Yazdi, ZN, ed. Ilahdad,

II, 100/ed. Urunbaev, fol. 320b, gives 40,000 foot, and does not at this stage specify the number of

elephants. But later we are told that 120 were captured: ed. Ilahdad, II, 118/ed. Urunbaev, fol. 325a.

122 cAfif, 144; but for the first expedition, cf. 115 (three divisions of 30,000 horse each), and

Hodivala, Studies, II, 123-4. 123 cAfif, 197, 200.

124

Ibid., 298. For cAfif's father, see ibid., 197. Digby, War-horse, 24-5.

125 TMS, 157. Digby, War-horse, 64, 76-7.

126 cAfif, 23.

would no doubt have played a part. Nor should we discount extraneous factors operating on the supply of warhorses. At the time of Ibn Battuta's travels, the lands of the Golden Horde had exported fine

mounts in droves of around 6000; but their availability would almost certainly have been considerably

reduced by the struggles among numerous rival Jochid khans since 759/1358.127 It is significant that no

Sultanate coins later than Firuz Shah's reign have been found in hoards from Russia.128

To what extent can the decline in military effectiveness be related to economic conditions? The

reputation Firuz Shah's reign acquired for widespread prosperity (above, pp. 169-70) seems to have been

derived from two closely related circumstances: a restoration of agrarian productivity following the death

of Muhammad b. Tughluq, and a fall in the price of grain and many other commodities. As far as the first is

concerned, Firuz Shah's personal efforts to promote cultivation are well known. The several canals that he

caused to be excavated transformed traditional areas of pasture into flourishing agricultural land.129 cAfif

devotes space especially to the two canals that irrigated the territory of the sultan's new foundation of Hisar Firuza, making a spring crop possible for the first time in addition to the autumn crops that had traditionally

been harvested in the region.130 Steps were also taken to bring waste land under the plough and to restore

the settlements that were attached to pious foundations like the tombs of shaykhs and past sultans.131 A

hundred thousand bighas of waste land were made over to faqirs and the needy.132

Yet the agrarian recovery does not seem to have brought in its wake a revival of the military

strength that had characterized the first decade of Muhammad b. Tughluq's reign; and two reasons appear to

have been a reduction in the government's revenues, and an increase in expenditure on builDing works and

for charitable purposes. There is some evidence that by this time the land-tax (kharaj) had been reduced to

20 - or even 10 - per cent.133 In c. 759/1358, moreover, on the basis of a tour of the empire by Husam al-

Din Junaydi, the gross revenue-demand was fixed at six krors and seventy-five laks (67,500,000) of tangas, and it remained at that level throughout, with the result that the government failed to benefit from enhanced

production in the provinces.134 cAfif was told, too, that the

127 On the horse trade, see IB, II, 372-4 (tr. Gibb, 478-9); Digby, War-horse, 35-6.

128 Digby, 'Currency system', 100; A. A. Bykov, 'Finds of Indian medieval coins in east Europe',

JNSI 21 (1965), 151-5.

129 TFS, 566, 567-71. SFS, 74-5, 161ff., 216-17.

130 cAfif, 127-8.

131 Ibid., 130, 332-3.

132 Ibid., 179: for the correct reading of this sentence, see Hodivala, Studies, II, 129-30.

133 Riazul Islam, 'Some aspects of the economy of northern South Asia during the fourteenth

century', JCA 11, part 2 (1988), 9 and n.21 (citing Mutahhar). But cf. cAfif, 484, yaklba-dah.

134 Ibid., 94, saying that Junaydi toured the empire for six years; at 296 the figure given is six krors

and eighty-five laks. Riazul Islam, 'Some aspects', 17-18.

abolition of uncanonical taxes in 777/1375-6 cost Flruz Shah thirty laks (3,000,000) of tangas.135 At the same time, the sultan is said to have set aside a total of 3,600,000 tangas for the culama', shaykhs

and holy men.136 A letter of Ibn Mahru provides some insight into conditions in the Multan province,

probably within a few years of Firuz Shah's accession. Answering the criticism, among others, that he had

been assigning unproductive land by way of pensions and allowances, Ibn Mahru draws attention to the fact

that the abolition of mukus under Muhammad b. Tughluq, and Firuz Shah's failure to reinstate them, has

reduced the sultan's revenue; and points at the same time to Firuz Shah's generosity in allocating an

unprecedentedly high sum of 300,000 tangas to the payment of pensions and gifts.137

Ibn Mahru's letter throws into sharp relief another problem confronting the government. The value

of stipends and pensions that had been fixed in kind at a time of high grain prices had been severely reduced when those prices fell - from as much as eighty jitals to a mere eight jitals per mann, if Ibn

Mahru's figures are reliable. Even without the sultan's partiality for sayyids, shaykhs and other deserving

causes, it would therefore have been deemed necessary to raise the grain-price equivalent of such grants in

order to protect the recipients against hardship.138 On the other hand, the sharp inflation of other prices

consequent upon Muhammad's debasement of the currency had not gone into reverse, so that the

government was confronted with a much higher bill for the purchase of essential war-material. The price of

horses, for instance, appears to have risen six- to eightfold since the time of Ala'al-Din KhaljI.139

Thus the sums available for expenditure on the military had undergone a reduction on several

counts. It was perhaps for this reason that Firuz Shah - at an early date, since it is mentioned by Barani -

had reverted to the policy of paying the regular troops in assignments of land; and cAfif may be referring to

this when he claims that the sultan gave away his whole empire in iqtacs.140 The soldiers in question he terms wajhdars, as opposed to those (ghayr-wajhis) who received pay either in cash or in drafts (barat) on

provincial revenue. cAfif, commenting on cAla' al-Din Khalji's refusal to follow such a practice on the

grounds that it created entrenched local interests, gives it as his own opinion that nevertheless no ill effects

could be detected during the forty years of Firuz Shah's reign.141 Yet short-term problems can certainly be

discerned. One distinction between the two types of trooper was that the wajhdars were expected to provide

their own

135 cAfif, 378-9.

136 Ibid., 179. TFS, 559, refers simply to an increase in the sum disbursed on pensions.

137 IM, 79-80.

I38 Ibid., 74.

139 Digby, War-horse, 37-40.

140 cAfif, 94-5, 279. TFS, 553.

141 cAfif, 96. On the two different types of trooper, see also ibid., 193-4, 296; Hodivala. Studies, I,

321-2.

mounts, which put them at a disadvantage as compared to the ghayr-wajhis and caused no little hardship among them at the time of the sultan's retreat from Thatta to Gujarat, when most of the horses had

been lost: their assignments being far away, it was necessary to advance them loans from the treasury.142

But there were also unwelcome longer-term effects. Troopers presenting their drafts in the iqtacs, cAfif tells

us, received only half the sum to which they were entitled. In these circumstances, many of them were

prepared to sell their drafts in Delhi for one-third of the total payment due, thus sparing themselves the

effort and expense of travelling to the iqtac. A brisk traffic thus developed in the drafts for soldiers' pay, and

many persons became wealthy by buying drafts at one-third of the nominal value and receiving in the

locality fifty per cent.143

At the same time as the introduction of pay through assignments, the sultan had also enacted

another measure permitting the wajhdar to transmit his establishment (istiqamat) to his son or son-in-law: failing them, it should pass to his slave or to some kinsman; and in the absence of these, lastly, to his

womenfolk (cawrat).144 The undesirable consequences of such a provision from the military vantage-point

are obvious. The complaint made to the sultan by Malik Ishaq, son of the "arid Bashir cImad al-Mulk, that

many of the troops had grown too old for service, echoes Barani's account of Balaban's attempt to change

the state of affairs in the hawali and the Doab soon after his accession (see p. 95). Firuz Shah ordered that

any soldier incapable of fulfilling his duties should provide a substitute (wakil).145 Nevertheless, the object

of the sultan's system may well have been to encourage exploitation of the land by giving each family a

permanent stake in the particular area allotted to it. The letter of Ibn Mahru cited above advocates giving,

as pay or pensions, a combination of cultivated and waste land.146

This would certainly have been in

keeping with the sultan's personal interest in extending cultivation.

But whatever the case, the number of Mahmud Shah's troops in 1398 must also give some idea of

the toll taken of the sultans' resources first by decades of mismanagement and then by some years of

internal conflict. In the early years of Firuz Shah's reign, Barani had commented on the new sultan's

indulgence towards the military.147 cAfif would be more outspoken, describing how Firuz Shah turned a

blind eye to the presentation of substandard horses and weapons at the annual review and retailing an

anecdote about the sultan's own efforts to help a trooper who had neglected to appear on time.148 Detailed

evidence is regrettably meagre; but we are left with the impression that the Sultanate's military

establishment had been run down. This trend can only have been accentuated by the internal strife and

142 cAfif, 220-1.

143 Ibid., 296-7.

144 Ibid., 96. The mention of women precludes the rendering of istiqamat as 'rank'.

145 Ibid., 302-3.

146 IM, 79-80.

147 TFS, 553. 148 cAfif, 298-302.

regional rebellion that characterized the years following the old sultan's death. cAfif writes in lyrical terms about the flourishing condition of the Doab under Firuz Shah;149 but within a few years much

of the province had been devastated by the campaigns of Hindu rais and rival Tughluqid princes.

The successor states

Sirhindi dates the emergence of autonomous provincial rulers from the time of the second civil

war, when, he claims, 'the amirs and maliks of the empire were independent sovereigns and would

appropriate the revenue and the produce themselves'.150 The process had in fact begun well before this, in

the reign of Firuz Shah, with the creation of an independent principality in Khandesh in 782/1380 under

Malik Raja, of whom we know little.151 It is a striking fact that apart from Malik Raja and with the qualified

exception of the creator of the Kalpi polity, Mahmud b. Firuz Khan, whose family had initially supported

Tughluq Shah II, the founders of the provincial dynasties152 - Khidr Khan at Multan, Zafar Khan Wajih al-Mulk in Gujarat, 'Amid Shah (Dilawar Khan) in Malwa, Shams Khan Awhadi at Bhayana and Khwaja

Jahan Sarwar at Jawnpur - were all originally nominees or supporters of Sultan Muhammad Shah b.

Firuz.153 Even Mahmud b. Firuz Khan must have made his peace with Muhammad Shah, who at the time of

his visit to the region in 794/1391-2 had conferred on him the iqtac of Mahoba in addition to the entire

shiqq of Firuzpur which he already held.154 Bihamadkhani goes so far as to equate the emergence of the

new kingdoms with an act of administrative convenience by Muhammad Shah, a formal division of his

territories into large administrative units following his triumph over Abu Bakr.155 The chronology of Malik

Mahmud's career alone suggests that we can take this story cum grano salis.

Khidr Khan at Multan is a special case. As we have seen, he had lost control of his province to

Sarang Khan, and doubtless no longer felt any loyalty - if he ever had done - to Sultan Mahmud Shah, who was a puppet of Sarang Khan's brother Mallu. Escaping from Sarang Khan's hands, he had fled to Bhayana,

and from there he made his way to Temur's encampment and offered his submission. Sirhindi's claim that

Temur bestowed Delhi upon him is surely apocryphal, an attempt to bolster the legitimacy of

149 Ibid., 295.

15° TMS, 160-1.

151

P. Hardy, 'Farukids', Enc.Isl2.

152 They are mostly listed in TMS, 168-9.

153 All referred to by Bihamadkhani, fols. 416a, 421b-422a, 426b (tr. Zaki, 19, 30, 36), except

Shams Khan, for whom see TMS, 147. Khidr Khan had also been an adherent of Muhammad Shah, ibid.,

146, 147.

154 Ibid., 152. Bihamadkhani, fols. 429b-430a (tr. Zaki, 42-3).

155 Ibid., fols. 426b, 429 (tr. Zaki, 36, 42). He may have been the source of the similar version in

Harawi, III, 288.

the Sayyid dynasty, for whom he was writing and who acknowledged Timurid overlordship, by means of the conqueror's imprimatur. His statement that Temur confirmed Khidr Khan as governor of

Multan and Deopalpur, however, we have no reason to doubt.156 In view of Khidr Khan's allegiance, these

territories had ceased to form part of the Delhi Sultanate.

Otherwise, however, the new rulers did not represent men who had come to power by any formal

act of rebellion. Khwaja Jahan Sarwar, wazir successively to Muhammad Shah and to Mahmud Shah and

viceroy to the latter throughout the eastern regions, is perhaps the most obvious case of a loyalist who

found autonomy thrust upon him. It is noteworthy, too, how hesitant these provincial governors were to

proclaim their own sovereignty and to repudiate the authority of the sultan in Delhi. At no time did Khwaja

Jahan assume the style of sultan; it was not until his death in 802/ 1399 (and therefore after the sack of

Delhi) that his adopted son and successor at Jawnpur took the title of Sultan Mubarak Shah, thereby provoking an abortive campaign by Mallu Iqbal Khan from Delhi.157 In Gujarat Zafar Khan b. Wajih al-

Mulk, despite Bihamadkham's statement to the contrary, displayed a reluctance to adopt the royal title

which is all the more surprising in one who had acknowledged Nusrat Shah and whose son Tatar Khan had

been that sultan's wazir. This would presumably explain Zafar Khan's embarrassment when Nusrat Shah's

rival Mahmud Shah appeared in Gujarat a year or so later, following Temur's invasion; the fugitive sultan

seems to have obtained no assistance and to have left for Malwa.158 In 806/1404 Zafar Khan, whom an

inscription of that year styles merely 'wazir', was briefly displaced by his ambitious son Tatar Khan, who

had designs on Delhi and adopted the title of Sultan Nasir al-Din Muhammad Shah. But even after Tatar

Khan's death and his own restoration two months later, Zafar Khan still called himself muqtac of Gujarat;

he did not take the title of sultan until 810/1407.159 It has been claimed that Dilawar Khan did so in

804/1401-2, after Mahmud Shah's visit to Malwa, but the only evidence for this appears to be an inscription

of 807/1405; the 156 TMS, 166-7. Yazdi, ZN, ed. Ilahdad, II, 175 (abridged in Urunbaev edn, fol. 341a), refers only

to the government of Multan.

157 TMS, 169. Harawi, III, 274. Saeed, Sharqi Sultanate, 32-3, says that around the time of Temur's

invasion he assumed the style of Atabeg al-Aczam and had the khutba read in his own name; but

Bihamadkhani, the source cited, does not support him. Cf. also Nizami, in HN, 713.

158 Bihamadkhani, fol. 427b (tr. Zaki, 38). TMS, 166, 170. Harawl, III, 89. Sikandar 'Manjhu', 20

(tr. Bayley, 79-80). Lai, Twilight, 47.

159 G. H. Yazdani, 'Seven new inscriptions from Baroda State', EIM (1939-40), 2-3. Z. A. Desai,

'Inscriptions of the Gujarat Sultans', EIAPS (1963), 6-10; idem, 'Khalji and Tughluq inscriptions from

Gujarat', 32-3, 38-40. TMS, 172. Harawl, III, 90-3, implying he took the royal title after Tatar Khan's death.

Sikandar 'Manjhu', 21-5 (tr. Bayley, 80, 81, 83-4). Misra, Muslimpower, 152-6.

evidence for the assumption of the royal title by his successor Hushang Shah is much stronger.160

The eventual assumption of sovereign status by Mahmud b. Firuz Khan, the founder of the

principality of Kalpi who reigned as Nasir al-Din Mahmud Shah, is a matter concerning which the local

chronicler, Bihamadkhani, is disarmingly confused. At one point he alleges that the Delhi Sultan Mahmud Shah b. Muhammad sent Mahmud b. Firuz Khan a chatr and a durbash, together with the title of sultan.

Slightly later, Mahmud is said to have established himself at Kalpi following the death of Mahmud Shah -

an impossible feat, since he died in 813/1410-11 and the Delhi Sultan survived for another two years. With

Bihamadkhani's statement immediately below, that Mahmud adopted the insignia of sovereignty in the

wake of Temur's invasion, we are doubtless as near to the truth as we shall get.161

Temur's assault on Delhi had been decisive. The artisans and other skilled workers who had

helped to beautify Firuz Shah's residences had been carried off to adorn the invader's headquarters at

Samarqand.162 Many of the city's other inhabitants had fled elsewhere for safety and had not returned.

Certainly, Bihamadkhani gives the impression that the security and prosperity of Kalpi were greatly

enhanced by the influx of refugees from Delhi in the wake of its sack by the Chaghadayid Mongols.163 The

collapse of the Delhi Sultanate was as much a matter of the death-blow to the capital city and its region as of the secession of most of its remaining provinces.

160 U. N. Day, Medieval Malwa: a political and cultural history, 1401-1562 (Delhi, 1965), 21;

HN, 898, 899. For the inscription, see EIM (1909-10), 11-12, summarized by Day, 435. For Hushang, see

ibid., 25.

161 Bihamadkhani, fol. 436 (tr. Zaki, 52, 53); and see also fol. 412b (tr. 15). For the death of

Mahmud Shah of KalpI, see ibid., fol. 445b (tr. 62).

162 Ghiyath al-Din Yazdi, Ruz-Nama, tr. Semenov, 124-5. Yazdl, ZN, ed. Ilahdad, II, 124/ed.

Urunbaev, fol. 326b. For other towns sacked, see Verma, Dynamics of urban life, 65-6. 163 BihamadkhanI, fols. 436b, 442b-443a (tr. Zaki, 53, 59-60). On the KalpI polity, see generally

Iqtidar Husain Siddiqui, 'Kalpi in the 15th century', IC 61 (1987), part 3, 90-120.

Epilogue (c. 1400-1526)

The end of the Tughluqids

During the decade following Temur's attack, the Sultanate reverted to being simply one of a

number of competing powers in the northern half of the subcontinent, as Firuz Shah's empire split into

several states. For three years after Temur's onslaught, there was not even a sultan in Delhi. Nusrat Shah,

who returned from the Doab to take up residence at Firuzabad in Rajab 801/March-April 1399, was defeated by Malm Iqbal Khan and obliged to flee into the Meo territory, where he died. Mallu then

established his headquarters at Siri, from where he is said to have brought back under control 'the shiqq of

the Doab and the iqtacs of the hawdli. But although he routed Sumer and his allies near Patiyali in

803/1401, he was unable to recover Gwaliyor from the successor of the Tomara chief Virasinha, who had

seized it during the chaos of Temur's onslaught.2 For a time Mallu was able to rule through Mahmud Shah,

whom he persuaded to rejoin him; but the sultan grew suspicious of him during a campaign against

Jawnpur and established himself at Qinnawj. He returned to Delhi only after Mallu's death in battle with

Khidr Khan in 808/1405, and maintained a shadowy authority there until his own death in 815/1412. After

the short reign of the leading amir Dawlat Khan, Khidr Khan finally obtained possession of Delhi.

The Sayyid and Lodi dynasties

The rulers of the so-called Sayyid (817-855/1414-1451) and Lodi (855-932/ 1451-1526)

dynasties3 presided over an empire that was a mere shadow of its former self and which continued to

fragment. The truncated Sultanate was surrounded, and sometimes threatened, by Muslim rivals like

Jawnpur,

1 TMS, 167-8.

2 Ibid., 169-70, 171-2.

3 For these dynasties, see generally K. A. Nizami, 'Sayyids', Enc.Isl2; idem, 'The Saiyids (1414-

51)', in HN, 630-63; idem, The Lodis (1451-1526)', ibid., 664-709; S. M. Imamuddin, 'Lodis', Enc.Isl. ; Lal,

Twilight.

Gujarat, Malwa and Bengal, and by renascent Hindu principalities in Mewar, Alwar and the Doab.

On occasions Delhi itself was menaced by invaders from one of the rival Muslim kingdoms, as it was by

the sultan of Malwa in 844/1440 and by the sultan of Jawnpur in 810/1407, in 856/1452, in c. 1466 and in

883/1479, just before the final overthrow of the Jawnpur Sultanate by Bahlul Lodi. Caliphal diplomas from

Cairo were now despatched to other Muslim monarchs in the subcontinent.4

The title 'Sayyids' applied to the dynasty of Khidr Khan (817-24/ 1414-21) is based on the descent

from the Prophet ascribed to them on inadequate grounds by Sirhindi. At no time did Khidr Khan assume sovereign status, preferring the title Rayat-i Ala ('exalted standard'). As befitted a ruler who owed his office

to Temur, he paid tribute to the conqueror's youngest son Shah Rukh, who now dominated the eastern

Islamic world from his capital at Herat, and was sent in exchange a robe of honour and a banner. And

although Sirhindi salutes Khidr Khan's son and successor Mubarak Shah (824-837/1421-1434) as sultan,5

we know that he too received from Herat a robe and a chatr.6 Sirhindi is silent on these contacts, and the

Sayyids' coinage did not bear Shah Rukh's name, comprising simply updated Tughluqid issues. We should

know nothing of the allegiance of the rulers of Delhi were it not for Bihamadkhani, who assures us that

Shah Rukh's orders had been received in Delhi for almost forty years and that the current ruler, Mubarak's

nephew Muhammad Shah (837-849/ 1434-45), was still obedient to him at the time of writing.7 The

subservience of Khidr Khan and his successors did not guarantee the Delhi Sultanate freedom from Mongol

attacks. Shaykh CA1i, who governed Kabul on behalf of Shah Rukh's son, profited from the Sayyids' difficulties to invade India on a number of occasions, briefly occupying Lahore in 836/1432-3.

Shah Rukh's influence in the subcontinent seems to have been extensive. Bihamadkhani, who

includes verses in praise of that monarch's sovereignty (saltanat) and refers to him as 'the seal of kings'

(khatam al-muluk),8 claims that Sultan Hushang Shah of Malwa appealed to him for assistance against an

invasion from Gujarat; while a Timurid source depicts the sultan of Bengal likewise seeking aid from Herat

against Jawnpur.9 This overlordship in all likelihood lapsed with the onset of civil war following Shah

Rukh's death in 850/1447 and the emergence of the threat to the Timurids from the Turkmens in western

Persia.10 But when Temur's descendant Babur

4 Otto Spies, 'Ein Investiturschreiben des abbasidischen Kalifen in Kairo an einen indischen

Konig', in S. M. Abdullah (ed.), Professor Muhammad Shaftc presentation volume (Lahore. 1955), 241-53. 5 TMS, 193.

6 Bihamadkhani, fols. 311b-312a (tr. Zaki, 95).

7 Ibid., fol. 312a (tr. Zaki, 95).

8 Ibid., fols. 312b-313a (not in Zaki's tr.).

9 Ibid., fol. 312b (tr. Zaki, 96). cAbd al-Razzaq Samarqandi, Matla al-Sacdayn, ed. M. Shaifi

(Lahore, 1941-9, 2 vols.), II, 782-3, cited in HN, 719. 10 See H. R. Roemer, The successors of Timur', in Jackson and Lockhart (eds.), Cambridge history

of Iran, VI, 105ff.

launched his five invasions of India from Kabul early in the sixteenth century, he was reviving the

claims of his forebears to sovereignty east of the Indus, although the conquest of Delhi itself became his

objective only with time.11

The early fifteenth-century sultans were barely able to impose their authority either on their own

muqta's or on local Hindu princes. Multan, once Khidr Khan's power-base, seceded in 847/1443 under Shaykh Yusuf Qurayshi, a descendant of Shaykh Baha' al-Din Zakariyya, who was subsequently

supplanted by the Afghan dynasty of the Langahs. Sirhindi's survey of the first decades of the Sayyid

dynasty amounts to little more than a tedious litany of campaigns against the Khokhars, the Meos, the

muqaddams of Katehr, the Chawhans of Etawa and the Tomaras of Gwaliyor, designed to raise 'revenue' in

the form of tribute payments. The sultanate of Delhi consisted of little more than the territories immediately

surrounding the capital itself, the hawali as they had long been known. One contemporary wag

immortalized by a sixteenth-century chronicler described the sway of the last Sayyid ruler, cAla' al-Din

'Alam Shah (shah-i 'alam, 'world-king'), as extenDing from Delhi to Palam.12

In pursuit of his designs on Delhi, Khidr Khan had recruited considerable numbers of Afghan

chiefs and their retinues. Already in his reign the Lodi chieftain Sultan Shah (later styled Islam Khan), who

had killed Mallu Iqbal Khan, held Sirhind; he fell in 834/1431 fighting against a Timurid invaDing force. During the 1440s Afghan nobles became the real power in the Sultanate. Islam Khan's nephew and

successor at Sirhind, Bahlul Lodi, who had been granted Lahore and Deopalptir in return for assistance

against the invaDing Malwa forces, went on to occupy most of the Panjab and made two attempts on Delhi. cAlam Shah abandoned the capital for Bada'un in 852/1448, and three years later Bahlul entered the city

and was enthroned as sultan.

The Lodi era witnessed something of a revival. A protracted duel with Jawnpur ended with its

annexation (884/1479); the region was later conferred on Sultan Bahlul's younger son Barbak Shah; the last

Sharqi sultan fled into Bihar. Here he maintained himself until his expulsion and the annexation of that

territory by Bahlul's son Sikandar Shah (894-923/ 1489-517), who had earlier removed his brother Barbak

from Jawnpur. Under Sikandar significant gains were also made to the south. The Awhadls, who had continued to rule Bhayana under the overlordship of Delhi, were finally ousted in 898/1492, when the place

was subjected to a nominee of the sultan. Narwar was wrested from the prince of Gwaliyor in 914/1508 and

Chanderi from the sultanate of Malwa in 921/1515; and in

11 Babur-Nama, tr. Beveridge, II, 377, 380, 382, 478.

12 Ahmad Yadgar, Ta'rikh-i Shahi or Ta'rikh-i Salatin-i Afaghina, ed. M. Hidayat Hosain, BI

(Calcutta, 1939), 5, cited in Lal, Twilight, 124 n.64.

915/1509 Nagawr became suborDinate to Delhi. Sikandar's son and successor, Ibrahim (923-

32/1517-26) succeeded in the conquest of Gwaliyor, which had eluded his father, but lost Chanderi and

Nagawr to the Hindu princes of Mewar and Marwar respectively. It is symptomatic of the sultans' preoccupation with the subjection of Alwar, Gwaliyor and Bhayana that Sikandar had in 911/1505 removed

his residence from Delhi to Agra. But the corollary of this forward policy in the south was neglect of the

vulnerable frontier in the Panjab.

The immigration of Afghan chiefs and their followers continued apace under the first two Lodis,

particularly when Bahlul, confronted by the threat from Jawnpur, sought to enhance his military strength by

inviting in tribesmen from the Roh clans.13 The position of the Afghan chiefs, and one or two non-Afghan

clans from the north-west with whom they shared power, was considerably stronger than that of their

precursors during the fourteenth century: it is noteworthy that the sultan around this time lost his long-

cherished monopoly of the elephantry.14 Bahlul ruled merely as primus inter pares. Sikandar, whose

ambitions were more autocratic, consolidated his position gradually and with tact; but Ibrahim from the outset showed himself to be uncompromising in his designs to curb the power of the older nobility and to

build up an elite upon which he could rely.15 His arbitrary actions against leaDing figures eventually

provoked the secession of Bihar under a rival, who seized the territory as far west as Qinnawj,16 and an

invitation from Dawlat Khan Lodi, governor of the Panjab, to the Timurid prince Babur to embark on his

last two invasions of India. In the fourth expedition, Lahore was occupied (930/1524), and in the fifth

Babur conquered the Delhi Sultanate. On 8 Rajab 932/20 April 1526, at Panipat, Ibrahim's superior

numbers were outclassed by Babur's artillery, and he fell in the fighting. Although the expulsion of Babur's

son Humayun, and the temporary establishment of a new Afghan-ruled polity by Shir Shah in 947/ 1540,

has some claim to be regarded as a recreation of the Delhi Sultanate, the engagement at Panipat marks the

beginning of the Mughal empire.

Babur is keen to contrast his own victory over the ruler of most of northern India with the

triumphs of the earlier conquerors, Mahmud of Ghazna and Mu'izz al-Din Muhammad of Ghur, whose

opponents had been smaller fry. In his opinion, Ibrahim's avarice was to blame for the fact that his army

stood at not more than 100,000 troops when he might have mustered twice or three times as many.17 But

this may not do justice to the

13 Ta'rikh-i Shir-Shahi, cited in HN, 679-80.

14 HN, 665. For the position of the Afghans in the Lodi Sultanate, see Iqtidar Husain Siddiqui-'The

composition of the nobility under the Lodi Sultans', MIM4 (1977), 10-66.

15 For the relations of the Lodi sultans with their nobles, see generally Iqtidar Husain Siddiqi.

Some aspects of Afghan despotism in India (Aligarh, 1969), chapters 1 -2.

16 Babur-Nama, tr. Beveridge, 523.

17 Ibid., 470, 480; see also his comment on the wealth of India in gold and silver, ibid., 518, 519.

absence of aid from the Lodi sultan's rebellious eastern provinces and a more widespread

alienation on the part of his army which is mentioned by a later source.18 It is also possible that Babur over-

estimated his enemy's wealth. The discontinuance of silver and gold coinage under the Sayyid and Lodi

sultans and the employment of baser metals such as billon and copper testify to the economic weakness of the fifteenth- and early sixteenth-century Sultanate, which had forfeited control of much of its land revenue

and no longer enjoyed access to the enormous sums gained in plunder or in tribute during the Khaljl and

Tughluqid eras.

18 Ahmad Yadgar, Ta'rikh-i Shahi, 96.

APPENDIX I

Juzjani's use of the word 'Turk'

Sometimes Juzjani employs the word 'Turk' as a general ethnicon, as when we read of the khan (or

khaqans) of 'the Turks': TN, I, 230, 231 (tr. 84, 85); cf. also I, 281 (tr. 194), for 'the cap of the Turks'. In its broadest sense, it could even embrace for Juzjani, as for other Muslim authors, the non-Turkish Qara-

Khitan and Mongols, as ibid., II, 94, 98 (tr. 900, 935). He also refers to the inhabitants of the regions lying

to the north and north-east of Lakhnawti, against whom Muhammad b. Bakhtiyar headed his disastrous

invasion, as 'Turks': ibid., I, 429 and n.4 (tr. 566, 567). The reason seems to be that their facial features

were thought to resemble those of the Turks: ibid., I, 427 (tr. 560); IB, IV, 216 (tr. Gibb and Beckingham,

869).

But there is also evidence that for Juzjani the word 'Turk' denoted a Turkish ghulam. The clearest

indication of this is that in his section on the last Ghaznawid Sultan he employs the phrase 'Turk and free'

(atrdk-u ahrdr): TN, I, 243 (Raverty's tr., 114, does not quite bring out the sense). At other points the

context usually suggests that the Turkish slave guards of, say, the Ghaznawids are in question: ibid., I, 234 (tr. 95, 97); also I, 230, 235, 236, 250, 251, 258, 286, 314 (tr. 83-4, 98, 100, 129, 131, 149, 180. 204-5,

282). By contrast, the word 'Turk' is applied to nomadic Turkish groups far more sparingly. Juzjani

employs the term 'Turk' for only one free Turkish chieftain, the founder of the dynasty of the

Khwarazmshahs (although in reality he too was a ghulam): TN, I, 297 (tr. 233). When the Turkish nomads

of the steppe are not called specifically Seljiiks or Ghuzz, they are referred to as 'Turkmen' (e.g., II, 94), a

designation applied, for instance, to Seljuk himself: ibid., I, 213 (turkdn to read turkmanan, as in BL ms.,

fol. 93a), 245 (tr. 45, 116). Turkmen seems to be used by the twelfth-century writer Marwazi to denote

Turkish nomads who had accepted Islam: Tabai al-Hayawan, partial edn and tr. V. Minorsky, Sharaf al-

Zaman Tahir Marvazi on China, the Turks and India (London, 1942), Ar. text 18, tr. 29 (and notes at 94-5).

But cf. I. Kafesoglu, 'A propos du nom Turkmen'. Oriens 11 (1958), 146-50.

APPENDIX II

Qilich Khan Mascud b.

cAla' al-Din Jani

The widespread confusion regarding this important noble of the middle of the thirteenth century is

due to the vagaries of the Tabaqat-i Nasiri manuscript tradition and of the two printed editions by Nassau

Lees and by Habibi. Thus 'Qilich' sometimes occurs as 'Qutlugh', causing Nizami to identify him with

Balaban's great enemy Qutlugh Khan (HN, 262, 271-2); Nigam, 41, 198-9, 203, similarly confuses the two

men. The same form Qutlugh is also adopted by Aziz Ahmad, Political history, 245, 246, 258, although he

distinguishes Mascud-i Jani from Qutlugh Khan. Qilich Khan's full name can be determined from TN, I,

476 (tr. 673), where he is listed among the maliks of Nasir al-Din Mahmud Shah as Malik Jalal al-Din Qilich Khan-i Malik Jani (text reads XLJ; BL ms., fol. 188b, here has 'Toghril' in error), and I, 495 (tr. 712),

where he is called Malik Jalal al-Din Mascud Shah-i Malik Jani. At II, 35 (as in BL ms., fol. 208a), he is

called 'Qutlugh Khan son of Malik Jani' (see also Raverty's tr., 769). At II, 78 (as in BL ms., fol. 223),

however, he appears as Qilich (QLJ) Khan (cf. Raverty's tentative 'Qutlugh [Qulij]' at 848-9). The title is

from Tu. qilich, 'sword': Sauvaget, 'Noms et surnoms', no. 178.

APPENDIX III

Qara'unas and Neguderis

I have throughout accepted the identification of the Qara'unas with the Neguderis made by Aubin ('L'ethnogenese', 84-5), and do not intend to devote further space to the origins of this grouping. Our Indian

sources never refer to Neguderis, but they do occasionally employ the term Qara'unas. The following

examples are from two authors writing in the fourteenth century. In his account of the death of Balaban's

son Muhammad in battle with the Mongols in 683/1285, cIsami says that the prince was killed by a

Qara'una horseman. The word is misread in Usha's edition, 179, 180, as fuzuna (defined in his glossary as 'a

soldier not present at review and not entered on the muster-roll'), but the correct form is found in the

otherwise inferior text edited by Husain (Agra, [1938], 174, 175). Amir Khusraw too employs the term of

the Mongol warrior who was briefly his captor following the overthrow of Muhammad b. Balaban (GK,

IOL ms. 412, fol. 78b, with HRWNH in error; correct spelling in Bada'uni, I, 153). And in describing the

punishment meted out to Mongols captured during the invasion by Iqbal, Kopek and Taibu in c. 1306, he

says (KF, 46), 'And through the mingling of Qara'una and Mongol, there was seen in every fortress the

junction of Saturn and Mars.'

There are several word-plays in this sentence, which hinges on the double meaning of burj as

'tower' and 'sign of the Zodiac'. 'Qara'una' could be read also as qaruna, 'soul', and 'Mughal' as maghal,

'sleep'. It is possible, lastly, that in 'Saturn', used by Indian Muslim writers to denote the infidel Hindu, we

have an allusion to the mixed Mongol-Indian descent of the Qara'unas - assuming, of course, that Marco

Polo's definition (guasmil. 'half-breeds') is reliable (Aubin, 'L'ethnogenese', 66-9); but this is a matter of

conjecture.

Together with the evidence of IB, III, 201 (tr. Gibb, 649), who heard from Shaykh Rukn al-Din of

Multan that the Qara'unas were 'Turks' who 'dwelt in the mountains between Sind and the Turks [i.e.

Transoxiana and Turkestan]', these examples suggest that the term Qara'unas was widely current in Muslim India and that the term Neguderis was used only by the Mongols themselves and by authors writing in

Mongol Persia.

APPENDIX IV

cAyn al-Mulk Multani and

cAyn al-Mulk Ibn Mahru

Some confusion has arisen between cAyn al-Mulk Multani, who conquered Malwa for 'Ala' al-Din

Khaljl, and cAyn al-Mulk Ibn Mahru, who governed Multan under Muhammad b. Tughluq and Firuz Shah

successively and whose correspondence has come down to us. For the equation of the two men, see cAbdu'l Wali, 254-5; 'Abdur Rashid's introduction to IM, Iff.; Lal, History of the Khaljis, 340; Husain, Tughluq

dynasty, 80-1, 87, etc., and index; Nigam, 13, 18, 79, 82, 88, 158-9, 171, 173, 174, 179 (though

distinguishing them in the index!); Nizami, On history and historians, 211-16, esp. 212 n.l, and in

Supplement to Elliot and Dowson's History of India, III, 64-5; Conermann, Beschreibung Indiens, 163-4.

By contrast, B. P. Saksena, in HN, 615 n.67, and I. H. Siddiqui, 'cAyn al-Mulk Multani', Enc. Isl2,

Supplement, 104-5, make them two separate individuals.

cAyn al-Mulk Ibn Mahru is known to have been an Indian: IB, III, 344 (tr. Gibb, 722). His

patronymic, for which see ibid., III, 342 (tr. 721), probably indicates that his father had been a convert to

Islam. His name appears in fuller form twice in his correspondence. On the first occasion, the diploma

appointing him to Multan calls him 'Malik al-Sharq wa'1-Wuzara' cAyn al-Mulk cAyn al-Dawla wa'l-Din ... cAbd-Allah-i Mahru' (IM, 12; cf. also SFS, 154, "Ayn al-Mulk cAyn al-Din-i Mahru'). Later, the author refers to himself as "Abd-Allah-i Muhammad Sharaf, known as (al-mad'u ba-) cAyn-i Mahru' (IM, 176).

This appears to preclude his identification with cAyn al-Mulk Multani, whose full name is given as cAyn al-

Mulk Shihab-i Taj Multani in TMS, 77, 87, and who is not heard of after Tughluq Shah's reign. In Tughluq-

Nama, 67, this earlier cAyn al-Mulk is made to claim Muslim ancestry as far back as ten generations (ba-

dah pusht), which suggests that he belonged to an immigrant Muslim family. The 'Malik Nasir al-Din, son

of cAyn al-Mulk', who according to IB, IV, 45 (tr. Gibb and Beckingham, 793), died when taking part in an

attack on Sindapur (Goa), would have been his son, since we are told that he lived at Ujjain.

APPENDIX V

The date of Ghiyath al-Din Tughluq Shah's death

Ghiyath al-Din Tughluq's death at Afghanpur and the accession of Muhammad have traditionally

been placed by historians in 725, in accordance with the date given in TFS, 456, for the latter event and the

month Rabi I of that year (February-March 1325) supplied for the Afghanpur episode by TMS, 96-1. But

the sources are far from unanimous. cIsami -like Barani, a contemporary (though admittedly far from

reliable as regards dates) - places Tughluq's death in 724 (FS, 421), and Husam Khan specifies the last day

(salkh) of that year (AHG, III, 862). Our data on the duration of Tughluq's reign are similarly vague. cAfif,

41, puts it at four and a half years; Barani at one point gives 'four or five years' (TFS, 438), but more often

puts it at 'four years and some months' (ibid., 22, 445); in this he is followed by Sirhindi, although in one

ms. chand is amended to hasht (TMS, 97 and n.l). But Baram's first recension, in this respect a hitherto

untapped source, furnishes a more exact figure of 'four years and four months' (TFS1, Bodleian ms., fol.

lla). Now Amir Khusraw dates Tughluq's accession on 1 Sha'ban 720/6 September 1320 (Tughluq-Nama, 135), and the figure in TFS1 would put his death somewhere in Dhu'l-Hijja 724 - in other words, at the very

end of the year, as indicated by Husam Khan. This conclusion is supported by a farman of Muhammad b.

Tughluq, dated 14 Dhu'l-Hijja 724/2 Dec. 1324, in which he is clearly the ruling sultan and his father is not

mentioned (Nizami, 'Some documents', 308-9). It is also very probable that an inscription on Firuz Shah's

column at Fathabad, in which Tughluq's death is dated Rajab 725 and Muhammad's accession on 1 Sha'ban

is in error by a whole year, given that this same epigraph sets the sultan's reign at four years and two

months (i.e. middle of 720-late 724): for the text, see Shokoohy, Haryana I, 21 and Pis. 28a, 29b-e; and cf.

review by Jackson, JRAS (1990), 171-2.

We further possess an inscription of Muhammad, as sultan, from Kanbhaya dated 18 Muharram

725/4 January 1325: this was edited by Husain ('Six inscriptions', 29-33), who goes to great lengths to prove that this date fell within Tughluq's reign and that it therefore applies to the commencement of the

building, which must have been completed several

months later. An inscription from Batihagarh, north-west of Damoh, admittedly bears the date 725

and names Tughluq as the reigning sultan (Verma, 'Inscriptions from the Central Museum, Nagpur', 111-

12). Nevertheless, since no month is given, it possibly belongs to the beginning of the year. The balance of

the evidence seems to be that the sultan died at the very end of 724, and I have accorDingly adopted this

date.

APPENDIX VI

The ancestry of Tughluq Shah II

Tughluq Shah II is everywhere called the son of Fath Khan. Although both Bihamadkhani, fol.

416a (tr. Zaki, 19), and TMS, 140, also call him Firuz Shah's grandson, it seems that this is an error. He was

in reality the old sultan's great-grandson, and the conventional genealogy of the later Tugh-luqids (e.g. in

Haig, Cambridge history of India, III, 189, 692; Banerjee, History of Firuz Shah, 47; Lai, Twilight, 2)

stands in need of emendation. Firuz Shah had four sons, Firuz Khan (known as shahzada-yi buzurg, 'the

great prince'), Zafar Khan, Muhammad Khan (the future sultan) and Shadi Khan, as listed in TFS, BL ms.,

fol. 260b (the phrase that follows in the printed text, 527, is corrupt and omits the two lastnamed princes),

and in Bihamadkhani, fol. 416b (tr. Zaki, 20). Fath Khan is explicitly referred to as Firuz Khan's son both

by Bihamadkhani (ibid.) and in TFS, 527 (the phrase acni sultan muhammad is a later interpolation, applying to Muhammad Khan, and has become displaced); cf. also 'Afif, 65, where Fath Khan is said to

have been born in Firuz Khan's house. Since his birth occurred in 752/1351 (TMS, 122), Fath Khan could

easily have had a young son by the time of his death - the sources comment on Tughluq Shah's youth:

Bihamadkhani, fols. 418a, 419b (tr. Zaki, 25, 27); TMS, 142. It has helped to confuse matters that Fath

Khan was in fact virtually a year older than his uncle Muhammad, for whose birth, on 3 Jumada I 753/17

June 1352, see TMS, 123.

Glossary

akhurbegl amir-i akhur intendant of the stables

amir-hajib military chamberlain

amir-i dad military justiciar

amir-i majlis intendant of the private assembly

amir-i sada commander of a unit of 100

amir-i shikar intendant of the hunt

'arid muster-master

Barbeg = amir-hdjib

Band intelligence officer; spy

chashnigir cupbearer

Chart ceremonial parasol chawdhuri Hindu chief/official in charge of a district

Dadbeg = amir-i dad

dhimma status of dhimmi

dhimmis 'Protected peoples' living under Islamic rule

diwan-i wizarat imperial revenue ministry

durbash ceremonial baton farrash palace attendant (literally 'carpet-spreader')

fath-nama victory despatch

ghayr-wajhis troops paid other than by assignments of land (see

pp. 316-17)

ghulam slave

Hajib chamberlain

Hawali territory in the environs of Delhi

in 'am (revenue grant) exempt from service

Insha correspondence; the art of prose composition

iqtac transferable revenue assignment in lieu of salary

Jizya capitation tax imposed on non-Muslims

karkhana manufactory, workshop

khalisa crown lands khanaqah sufi hospice

kharaj land-tax; tribute; (more generally) revenue

kharitadar keeper of the purse

khass-hdjib privy chamberlain

khutlkhot (Hindu) headman

kotwal Castellan

kror 100 laks, i.e. 10 million

kuroh approximately 2 miles

lak 100,000

mawas (Hindu) territory inaccessible to Muslim attack (see

p. 125)

mawlazada son of a freed slave

muhrdar keeper of the seal

muhtasib overseer of public morality; inspector of the markets

Mucus taxes not sanctioned by the Sharfa

muqaddam (Hindu) chief

Muqta' holder of an iqtac

mushrif-i mamalik accountant-general of imperial revenue

mustawfi-yi mamdlik auditor-general

mutasarrif (provincial) revenue-collector

na'ib [-i] viceroy; (deputy-)

na'ib-i card deputy muster-master

nawbat band playing outside royal or noble residence as a

mark of honour

Noyan (Mongol) commander

Paik (Hindu) infantryman

pilkhana elephant-stable

qadi-yi lashgar judge of the army

quriltai (Mongol) assembly of princes and generals

Sadi hundred (administrative division)

Sdh Hindu banker/moneylender

sar-i chatrddr chief parasol-bearer

sar-i dawatdar chief inkwell-holder

sar-i jandar commander of the sultan's guards or executioners

sar-i sildhdar chief armour-bearer

shihna governor; (Mongol) resident at the court of a subject

ruler

shihna-yi bargah intendant of the audience-hall

shihna-yi manda intendant of the markets

shihna-yi pil intendant of the elephantry

Shiqq administrative division

sildhddr armour-bearer

talwaraltalwandi (Hindu) territory or encampment

Tumen (Mongol) military unit of 10,000

Ulus complex of people, livestock and grazing grounds allotted

to a prince of the Mongol imperial dynasty

wajhdars regular troops paid in assignments of land (see pp. 316-17)

wakil-i dar comptroller of the household

wall governor

Select bibliography

Primary sources

(a) Literary sources

Anonymous, Baydd-i Taj al-Din Ahmad Wazir, ed. Iraj Afshar and Murtada Taymuri (Isfahan,

1353 Sh./1974)

Anonymous, continuation of Rashid al-Din, Jamic al-Tawarikh, Istanbul ms. Nuruosmaniye

Kutuphanesi 2799 (old numbering: 3271)

Anonymous, al-Hawadith al-Jami'a (wrongly attributed to Ibn al-Fuwati), ed. Mustafa Jawad

(Baghdad, 1351/1932)

Anonymous, Hudud al-'Alam, facsimile edn by V. V. Bartol'd (Leningrad, 1930); tr. V. Minorsky,

2nd edn, GMS, ns, XI (London, 1970)

Anonymous, Mu'izz al-Ansab, BN ms. Ancien fonds persan 67

Anonymous, Sirat-i Firuz-Shahi, SOAS ms. 283116 (copy of ms. in Bankipore Library, Patna);

extracts tr. K. K. Basu, 'An account of Firoz Shah Tughluq', JBORS 22 (1936), 96-107, 265-74; 23 (1937),

97-112; K. K. Basu, 'Firoz Tughluq and his Bengal campaign', JBORS 27 (1941), 79-95; N. B. Roy,

'Jajnagar expedition of Sultan Firuz Shah', JRASB, Letters 8 (1942), 57-98; J. A. Page, 'A memoir on Kotla

Firuz Shah', MASI 52 (1937), 33-42; Sh. Abdur Rashid, 'Firoz Shah's investiture by the Caliph', MIQ 1

(1950), 66-71

Anonymous, tr. of Razi's Sirr al-Makhtuma, BN ms. Suppl. persan 384 Anonymous, Ta'rikh-i

Sistdn, ed. Malik al-Shu'ara Bahar (Tehran, 1314 Sh./1935); tr. L. P. Smirnova (Moscow, 1974)

cAfif, Shams-i Siraj, Ta'rikh-i Firuz-Shahi, ed. Maulavi Vilayat Hosain, BI (Calcutta, 1888-91)

Ahmad Yadgar, Ta'rikh-i Shahi or Ta'rikh-i Salatin-i Afaghina, ed. M. Hidayat Hosain, BI

(Calcutta, 1939)

Amir Hasan Dihlawi, Fawa'id al-Fu'ad, ed. M. Latif Malik (Lahore, 1386/1966) Amir Khusraw

Dihlawi, Dibacha-yi Ghurrat al-Kamal, ed. Sayyid Wazlr al-Hasan

cAbidi (Lahore, 1975); IOL Persian ms. 51 (Ethe, no. 1186)

Diwal Rani-yi Khadir Khan, ed. Rashid Ahmad Salim Ansari (Aligarh, 1336/1917)

Ghurrat al-Kamal, IOL Persian ms. 412 (Ethe, no. 1187)

Khaza'in al-Futuh, ed. M. Wahid Mirza, BI (Calcutta, 1953)

Miftah al-Futuh, ed. Sh. Abdur Rashid (Aligarh, 1954)

Nuh Sipihr, ed. M. Wahid Mirza (London, 1950)

Qiran al-Sacdayn, ed. Maulavi Muhammad Ismacil and Sayyid Hasan Barani (Aligarh, 1337/1918)

Rasa'il al-I'jaz, lithograph edn (Lucknow, 1876, 5 vols. in 2); IOL Persian ms. 570 (Ethe, no.

1219) and BL mss. Add. 16841, Add. 16842

Tughluq-Nama, ed. Sayyid Hashimi Faridabadi (Aurangabad, 1352/1933)

Tuhfat al-Sighar, IOL Persian ms. 412 (Ethe, no. 1187)

Wasat al-Hayat, IOL Persian ms. 412 (Ethe, no. 1187)

cAwfi, Sadid al-Din Muhammad b. Muhammad Bukhari, Jawami' al-Hikayat, 1 (preface), ed.

Muhammad Mucin, 2nd edn (Tehran, 1350 Sh./1971); BL mss. Or. 4392 and Or. 2676; BN ms. Anc. fonds

persan 75

Lubab al-Albab, ed. E. G. Browne and M. M. Qazwini (Leiden and London, 1903-6, 2 vols.)

Babur, Zahir al-Din Muhammad, Bdbur-Nama, tr. Annette S. Beveridge, The Babur-nama in

English (London, 1921-2; repr. 1969)

Bada'uni, cAbd al-Qadir, Muntakhab al-Tawarikh, ed. Maulavi Ahmad cAli, BI (Calcutta, 1864-9,

3 vols.)

Badr-i Chach, Qasa'id, lithograph edn by M. Hadi CA1i (Kanpur, n.d.); lithograph edn by M.

'Uthman Khan (Rampur, 1872-3, 2 vols.)

Baladhuri, Futuh al-Buldan, ed. M. J. De Goeje, Liber expugnationis regionum (Leiden, 1866); ed. S. al-Munajjid (Beirut, n.d.)

Barani, Diya', Fatawa-yi Jahandari, ed. Afsar Saleem Khan (Lahore, 1972)

Nact-i Muhammadi, RRL Persian ms. 1895; extract tr. in S. Nurul Hasan, 'Sahifa- i-Nact-i-

Muhammadi of Zia-ud-Din Barani', MIQ 1 (1950), 100-6

Ta'rikh-i Firuz-Shahl, ed. Saiyid Ahmad Khan, BI (Calcutta, 1860-2); BL ms. Or.

Tarikh-i Firuz-Shdhl, first recension: Bodleian ms. Elliot 353; RRL Persian ms. 2053; ms. in

private collection of Simon Digby

Bihamadkhani, Muhammad, Ta'rlkh-i Muhammadi, BL ms. Or. 137; partial tr. (from 755/1354)

Muhammad Zaki (Aligarh, 1972)

Fakhr-i Mudabbir (Muhammad b. Mansur b. Sa'Id Qurashi), Addb al-Harb wa'l-Shajdca (or Addb

al-Muluk), ed. Ahmad Suhayll Khwansari (Tehran, 1346 Sh./ 1967)

Shajarat (or Bahr) al-Ansdb, partial edn by Sir E. Denison Ross, Ta'rikh [sic]-/ Fakhr al-Din

Mubdrakshah (London, 1927)

Firishta (Muhammad Qasim Hindu Shah Astarabadi), Gulshan-i Ibrdhlml, lithograph edn

(Bombay, 1247/1831-2, 2 vols.)

Firuz Shah b. Rajab (Sultan), Futuhdt-i Firuz-Shdhl, ed. Sh. Abdur Rashid (Aligarh, 1954); tr. N.

B. Roy, 'The victories of Sultan Firuz Shah of Tughluq dynasty', IC 15 (1941), 449-64

Ghaznawi, cAbd al-Hamld Muharrir, Dastur al-Albdb fi cIlmi'l-Hisdb, RRL Persian ms. 1231;

partial tr. Sh. Abdur Rashid, 'Dastur-ul-Albab fi cIlm-il-Hisab', MIQ 1 (1950), 59-99

Hamid Qalandar, Khayr al-Majdlis, ed. K. A. Nizami (Aligarh, [1959])

Harawi, Nizam al-Din Ahmad, Tabaqdt-i Akbarl, ed. B. De et al., BI (Calcutta, 1931-5, 3 vols.)

Hasan-i Nizami (Taj al-Din Hasan b. Nizami NIshapurl), Taj al-Ma'dthir, IOL Persian ms. 15

(Ethe, no. 210)

Ibn Abi'l-Fada'il, al-Mufaddal, al-Nahj al-Sadld, partial edn and tr. E. Blochet.

'Moufazzal Ibn Abil-Fazail. Histoire des Sultans Mamlouks', Patrologia Or ientalis 12-20 (1919-

29) (references are to the separatum); partial edn and tr. Samira Kortantamer, Agypten und Syrien zwischen

1317 und 1341 IU 23 (Freiburgi. Br., 1973)

Ibn al-Athir, 'Izz al-Din, al-Kamil fi'l-Ta'rikh, ed. C. J. Tornberg, Ibn-El-Athiri Chronicon quod

perfectissimum inscribitur (Leiden, 1851-76, 12 vols.); repr. (with different pagination) Daral-Sader (Beirut, 1386/1966, 12 vols.) (references are to both editions, in that order)

Ibn al-Dawadari, Abu Bakr b. 'Ali b. Aybak, Kanz al-Durar wa-Jamic al-Ghurar, ed. Said 'Ashur

et al, Die Chronik des Ibn ad-Dawadari, QGIA, I: VII (Cairo, 1392/1972); VIII (Cairo, 1391/1971); IX

(Cairo, 1379/1960)

Ibn Battuta, Abu cAbd-Allah Muhammad b. cAbd-Allah al-Lawati al-Tanji, Tuhfat al-Nuzzar ft

Ghara'ibi'l-Amsar, ed. Ch. Defremery and B. S. Sanguinetti (Paris, 1853-8, 4 vols.); tr. H. A. R. Gibb, The

travels of Ibn Battuta A.D. 1325-1354, HS, 2nd series, 110, 117, 141, 178 (Cambridge and London, 1958-

94, 4 vols. Of 5 so far; vol. IV ed. by C. F. Beckingham)

Ibn Mahru,cAyn al-Mulk cAbd-Allah-i Muhammad Sharaf, Insha-yi Mahru, ed. Sh. Abdur Rashid

(Lahore, 1965)

Ibn Said al-Maghribi, Kitab al-Jughrafiyya, ed. Ismail al-cArabi (Beirut, 1970)

Ibn Zarkub, Mucin al-Din Ahmad b. Abi'l-Khayr Shirazi, Shiraz-Nama, ed. Bahman Karimi

(Tehran, 1310 Sh./1932); ed. Ismail Wa'iz Jawadi (Tehran, 1350Sh./1971)

Ikhtisan-i Dabir (Taj al-Din Muhammad b. Ahmad b. Hasan), Basatin al-Uns, BL ms. Add. 7717

cIsami, cAbd al-Malik, Futuh al-Salatin, ed. A. S. Usha (Madras, 1948); tr. A. M. Husain (Aligarh,

1967-77, 3 vols. with continuous pagination)

Jafar b. Muhammad b. Hasan Ja'fari, Ta'rikh-i Yazd, ed. Iraj Afshar (Tehran, 1338 Sh./1959)

Jajarmi, preface to his tr. of Ghazali's Ihya "Ulumi'l-Din, BL ms. Or. 8194

Jamal al-Qarshi, Mulhaqat al-Surah, ed. in V. V. Bartol'd, Turkestan v epokhu mongol'skogo

nashestviia (St Petersburg, 1898-1900, 2 vols.), I (texts)

Juwayni, cAla' al-Din Ata Malik, Ta'nkh-i Jahan-Gusha, ed. Mirza Muhammad Qazwini, GMS,

XVI (Leiden and London, 1912-37, 3 vols.); tr. J. A. Boyle, The history of the world-conqueror

(Manchester, 1958, 2 vols. with continuous pagination)

Juzjani, Minhaj al-Din Abu cUmar cUthman b. Siraj al-Din, Tabaqat-iNasiri, ed. cAbd al-Hayy

Habibi, 2nd edn (Kabul, 1342-3 Sh./1963-4, 2 vols.); tr. H. G. Raverty, Tabakat-i Nasiri: a general history

of the Muhammadan dynasties of Asia, BI (London, 1872-81, 2 vols. with continuous pagination); BL ms.

Add. 26189; IOL ms. 3745; Berlin ms. Petermann 386 (Pertsch, no. 367)

Kirmani, Muhammad b. Mubarak (Mir-i Khwurd), Siyar al-Awliya fi Mahabbati'l- Haqq jalla wa-

'Ala, lithograph edn (Delhi, 1302/1885)

Kirmani, Nasir al-Din, Simt al-'Ula li'l-Hadrati'l-'Ulya, ed. cAbbas Iqbal (Tehran,1328Sh./1949)

Kufi, CA1I b. Hamid b. Abl Bakr, Chach-Ndma, ed. N. A. Baloch, Fathndmah-i Sind (Islamabad,

1403/1983)

Maqrizi, Taqi al-Din Ahmad b. CA1i al-, al-Suluk li-Macrifat Duwali'l-Muluk, ed. M.M. Ziada et

al. (Cairo, 1934- in progress)

Marco Polo, Le divisament dou monde, tr. A. C. Moule and Paul Pelliot, The description of the

world (London, 1938, 2 vols.); tr. Sir Henry Yule, The book of Ser Marco Polo the Venetian, new edn by

H. Cordier (London, 1920, 3 vols.)

Nasawi, Muhammad b. Ahmad b. cAli, Sirat al-Sultan Jalal al-Din, ed. Octave Houdas (Paris, 1891); tr. Z. M. Buniiatov, Zhizneopisanie Sultana Dzhalal ad-Dina Mankburny (Baku, 1973)

Natanzi, Mu'in al-Din, Muntakhab al-Tawarikh, partial edn by Jean Aubin, Extraits du Muntakhab

al-tawarikh-i Mu'ini (Tehran, 1957)

Nizami, K. A. (ed.), 'Some documents of Sultan Muhammad bin Tughluq', MIM 1 (1969), 305-13

Qashani, Jamal al-Din cAbd-Allah b. 'Ali, Ta'rikh-i Uljditu Sultan, ed. Mahin Hambly (Tehran,

1348 Sh./1969)

Qazwini, Hamd-Allah Mustawfi, Ta'rikh-i Guzida, ed. cAbd al-Husayn Nawa'I (Tehran, 1339 Sh./1960)

Rashid al-Din Fadl-Allah al-Hamadani, Jami' al-Tawirikh, divided as follows:

A. History of the Mongols:

I, part 1 [account of the Turkish and Mongol tribes], ed. A. A. Romaskevich et al. (Moscow,

1965); tr. A. A. Khetagurov, Sbornik letopisei, I, part 1 (Moscow and Leningrad, 1952); I, part 2 [history of

Chinggis Khan], ed. I. N. Berezin, 'Sbornik letopisei', TVOIRAO 13 (1868), 1-239, and 15 (1888), 1-231; tr.

O. I. Smirnova, Sbornik letopisei, I, part 2 (Moscow and Leningrad, 1952);

II [history of Chinggis Khan's successors], ed. E. Blochet, GMS, XVIII, part 1 (Leiden and London, 1911); tr. J. A. Boyle, The successors of Genghis Khan (New York, 1971); tr. Iu.P. Verkhovskii,

Sbornik letopisei, II (Moscow and Leningrad, 1960); II, part 1 (only), ed. A. A. Alizade (Moscow, 1980);

III [history of the Ilkhans], ed. A. A. Alizade and tr. A. K. Arends (Baku, 1957); partial edn by

Karl Jahn, Geschichte Gazan Han's, GMS, ns, XIV (London, 1940);

B. History of India, ed. and tr. Karl Jahn, Die Indiengeschichte des Rasid ad-Din, 2nd edn

(Vienna, 1980)

Shucab-i Panjgana, TSM ms. III Ahmet 2937

Safadi, Khalil b. Aybak al-, A'yan al-cAsr, SK ms. Asir Efendi 588

al-Wafi bi'l-Wafayst, ed. H. Ritter et al. (Damascus, 1931 —in progress); extract tr.M. S. Khan,

'An undiscovered Arabic source of the history of Sultan Muhammad bin Tughlaq', IC 53 (1979), 187-205,

with addenda (comprising Arabic text), IC 54 (1980), 47-8

Sayfi (Sayf b. Muhammad b. Ya'qub al-Harawi), Ta'rikh-Nama-yi Harat, ed. M. Z. as-Siddiqui

(Calcutta, 1944)

Sayyid Ashraf Jahangir Simnani, Maktubat-i Ashrafi, BL ms. Or. 267

Shabankara'i, Muhammad b. 'Ali, Majma" al-Ansab, ed. Mir Hashim Muhaddith (Tehran, 1363

Sh./1984)

Shami, Nizam-i, Zafar-Nama, ed. Felix Tauer, Histoire des conquetes de Tamerlan intitulee

Zafarnama, Monografie Archivu Orientalniho, V (Prague, 1937-56, 2 vols.)

Sikandar b. Muhammad, alias Manjhu, Mir'dt-i Sikandari, ed. S. C. Misra and Muhammad Lutf

al-Rahman (Baroda, 1961); tr. in Sir E. C. Bayley, The local Muhammadan dynasties. Gujarat (London,

1886)

Sirhindi, Yahya b. Ahmad, Ta'rikh-i Mubarak-Shahi, ed. S.M. Hidayat Husain, BI (Calcutta, 1931)

Ulughkhani, cAbd-Allah Muhammad al-Makki, Zafar al-Walih bi-Muzaffar wa-Alih, ed. Sir E.

Denison Ross, An Arabic history of Gujarat (London, 1910-28, 3 vols.)

cUmari, Shihab al-Din Abu'l-'Abbas Ahmad b. Yahya ibn Fadl-Allah al-, Masalik al-Absar fi

Mamaliki'l-Amsar, (section on India), ed. and tr. Otto Spies, Ibn Fadlallah al-cOmari's Bericht u'ber Indien

in seinem Werke..., Sammlung Orientalischer Arbeiten, XIV (Leipzig, 1943); ed. Khurshid Ahmad Fariq

(Delhi, [1961]); tr. Iqtidar Husain Siddiqi and Qazi Muhammad Ahmad, A fourteenth century Arab account

of India under Sultan Muhammad bin Tughlaq (Aligarh, [1975]); (section on the Mongol empire) ed. and

tr. Klaus Lech, Das mongolische Weltreich: al-cUmarl's Darstellung der mongolischen Reiche in seinem Werk..., AF, XXII (Wiesbaden, 1968)

Van den Wyngaert, Anastasius (ed.), Sinica Franciscana, I. Itinera et relationes Fratrum

Minorum saeculi XIII et XIV (Quaracchi-Firenze, 1929)

Wassaf (Shihab al-Din cAbd-Allah b. cIzz al-Din Fadl-Allah Shirazi), Tajziyat al-Amsar wa-

Tazjiyat al-A'sar, lithograph edn (Bombay, 1269/1853)

Yazdi, Ghiyath al-Din CAli, Ruz-Nama-yi Ghazawat-i Hindustan, tr. A. A. Semenov, Dnevnik

pokhoda Timura v Indiiu (Moscow, 1958)

Yazdi, Sharaf al-Din 'All, Zafar-Nama, ed. M. M. Ilahdad, BI (Calcutta, 1885-8, 2 vols.); facsimile (incluDing muqaddima) edn by A. Urunbaev (Tashkent, 1972)

Yunini, Qutb al-Din Musa b. Muhammad al-, al-Dhayl 'ala.' Mir'ati'l-Zaman (Hyderabad, AP,

1374 80/1954-61, 4 vols. so far); TSM ms. III Ahmet 2907/ e.3

Yusuf-i Ahl, Jalal al-Din, Fard'id-i Ghiyathi, ed. Hishmat Mu'ayyad (Tehran, 2536

Shahanshahi/1977 and 1358 Sh./1979, 2 vols. so far); SK ms. Fatih 4012

(b) Epigraphy and numismatics

Abdul Karim (ed.), Corpus of the Arabic and Persian inscriptions of Bengal (Dacca, 1992)

Ahmad, Q. (ed.), Corpus of Arabic and Persian inscriptions of Bihar (AH 640-1200) (Patna, 1973)

Buhler, G., 'A Jaina account of the end of the Vaghelas of Gujarat', IA 26 (1897), 194-5

Dani, Ahmad Hasan, Muslim inscriptions of Bengal, Appendix to JASP 2 (Dacca, 1957)

Desai, P. B., 'Kalyana inscription of Sultan Muhammad, Saka 1248', El 32 (1957-8), 165-70

Desai, Z. A., 'Inscriptions of the Mamluk Sultans of Delhi', EIAPS (1966), 4-18

'Khalji and Tughluq inscriptions from Gujarat', EIAPS (1962), 1 -40

'The Chanderi inscription of cAlau'd-Din Khalji', EIAPS (1968), 4-10

'The Jalor Idgah inscription of Qutbu'd-Din Mubarak Shah Khalji, EIAPS (1972), 12-19

'Two inscriptions of Ghiyathu'd-Din Tughluq from Uttar Pradesh', EIAPS (1966), 19-26

Diskalkar, D. B., 'Inscriptions of Kathiawad', NIA 1 (1938-9), 686-96

Husain, A. M., 'Six inscriptions of Sultan Muhammad bin Tughluq Shah', EIAPS (1957-8), 29-42

Moneer, Q. M., 'Two unpublished inscriptions of the time of Sultan Muhammad bin Tughluq',

EIM (1939-40), 23-6

Nazim, M., Inscriptions of Bijapur, MASI 49 (Delhi, 1936)

Prasad, Pushpa (ed.), Sanskrit inscriptions of Delhi Sultanate 1191-1526 (Delhi, 1992)

Shokoohy, Mehrdad (ed.), Haryana I, Corpus inscriptionum Iranicarum, part IV: Persian

inscriptions down to the early Safavid period, XLVII (London, 1988)

Rajasthan I, Corpus inscriptionum Iranicarum, part IV: Persian inscriptions down to the early

Safavid period, XLIX (London, 1986)

Shu'aib, M. M., 'Inscriptions from Palwal', EIM(1911-12), 1-5

Sircar, D.C., 'Inscriptions of the time of Yajvapala Gopala', EI 31 (1955-6), 323-36

Thomas, Edward, Chronicles of the Pathan kings of Delhi (Delhi, 1871)

Verma, B. D., 'Inscriptions from the Central Museum, Nagpur', EIAPS (1955-6), 109-18

Wright, H. Nelson, Catalogue of the coins of the Indian Museum, Calcutta, II (Oxford, 1907)

The coinage and metrology of the Sultans of Dehli (Delhi, 1936)

Yazdani, Ghulam Husain, 'Inscriptions of the Khalji Sultans of Delhi and their contemporaries in

Bengal', EIM (1917-18), 8-42

'The inscriptions of the Turk Sultans of Delhi - Mucizzu-d-Din Bahram, cAla'u-d-Din Mas'ud,

Nasiru-d-Din Mahmud, Ghiyathu-d-Din Balban and Mucizzu-d-Din Kaiqubad', EIM (1913-14), 13-46

Secondary sources cAbdu'l Wall, Maulavi, 'Life and letters of Malik 'Aynu'1-Mulk Mahru, and sidelights on Firuz

Shah's expeditions to Lakhnauti and Jajnagar', JASB ns 19 (1923), 253-90

Ahmad, Aziz, 'Mongol pressure in an alien land', CAJ 6 (1961), 182-93

Studies in Islamic culture in the Indian environment (Oxford, 1964)

'The early Turkish nucleus in India', Turcica 9 (1977), 99-109

'The sufi and the sultan in pre-Mughal Muslim India', Der Islam 38 (1962), 142-53

Ahmad, Laiq, 'Kara, a medieval Indian city', IC 55 (1981), 83-92

Ahmad, Muhammad cAziz, Political history and institutions of the early Turkish empire of Delhi

(1206-1290AD) (Lahore, 1949)

Ali, Athar, 'Military technology of the Delhi Sultanate (13-14th C.)', in PIHC 50 (Gorakhpur

1989) (Delhi, 1990), 166-82

Allsen, T. T., Mongol imperialism: the policies of the Grand Qan Mongke in China, Russia, and

the Islamic lands, 1251-1259 (Berkeley and Los Angeles, 1987)

'The Yuan dynasty and the Uighurs of Turfan in the 13th century', in Morris Rossabi (ed.), China

among equals: the Middle Kingdom and its neighbours, 10th-14th centuries (Berkeley and Los Angeles,

1983), 243-80

Askari, S. H., 'Material of historical interest in Ijaz-i-Khusravi', MIM 1 (1969), 1-20

'Risail (Sic)-ul-Ijaz of Amir Khusrau: an appraisal', in Dr. Zakir Husain presentation volume

(Delhi, 1968), 116-37

Aubin, Jean, 'L'ethnogenese des Qaraunas', Turcica 1 (1969), 65-94

'Le khanat de Cagatai et le Khorassan (1334-1380)', Turcica 8 (1976), 16-60

'Les princes d'Ormuz du XIIIe au XVe siecle', JA 241 (1953), 77-138 '

Ayalon, David, 'Aspects of the Mamluk phenomenon, I. The importance of the Mamluk

institution', Der Islam 53 (1976), 196-225; repr. in his The Mamluk military society (London, 1979)

Studies on the Mamluks of Egypt (1250-1517) (London, 1977)

'Studies on the structure of the Mamluk army - IF, BSOAS 15 (1953), 448-76; repr. in his Studies

'The Wafidiya in the Mamluk kingdom', IC 25 (1951), 81-104; repr. in his Studies

Banerjee, Jamini Mohan, History of Firuz Shah Tughluq (Delhi, 1967)

Barthold, W., Four studies on the history of Central Asia (Leiden, 1956-62, 4 parts in 3 vols.)

Turkestan down to the Mongol invasion, 3rd edn by C. E. Bosworth, GMS, ns, V (London, 1968)

Zwolf Vorlesungen u'ber die Turken Mittelasiens, tr. Th. Menzel (Hildesheim, 1935; repr. 1962)

Bhandarkar, D. R., 'The Chahamanas of Marwar', El 11 (1911-12), 26-79

Biran, Michal, Qaidu and the rise of the independent Mongol state in Central Asia (Richmond,

Surrey, 1997)

Bosworth, C. E., The Ghaznavids. Their empire in Afghanistan and eastern Iran 994:1040, 2nd

edn (Beirut, 1973)

The history of the Saffarids of Sistan and the maliks of Nimruz (247/861 to 949/ 1542-3) (Costa

Mesa, California, 1994)

The later Ghaznavids, splendour and decay: the dynasty in Afghanistan and northern India 1040-1186 (EDinburgh, 1977)

The medieval history of Iran, Afghanistan and Central Asia (London, 1977)

'The political and dynastic history of the Iranian world (A.D. 1000-1217)', in Boyle (ed.),

Cambridge history of Iran, V, 1 -202

Bouchon, Genevieve, 'Quelques aspects de l'islamisation des regions maritimes de l'lnde a

l'epoque medievale (XIIe-XVIe s.)', in Gaborieau (ed.), Islam et societe, 29-36

Boyle, J. A., 'Dynastic and political history of the Il-khans', in Boyle (ed.), Cambridge history of

Iran, V, 303-421 'The Mongol commanders in Afghanistan and India according to the Tabaqat-I Nasiri of Juzjani, IS 2 (1963), 235-47

Boyle, J. A. (ed.) The Cambridge history of Iran, V. The Saljuq and Mongol periods (Cambridge,

1968)

Chakravarti, Monmohan, 'Notes on Gaur and other old places in Bengal', JASB ns 5(1909), 199-

235

Clauson, Sir Gerard, An etymological dictionary of pre-thirteenth-century Turkish (Oxford, 1972)

Conermann, Stephan, Die Beschreibung Indiens in der 'Rihla' des Ibn Battuta: Aspekte einer herrschaftssoziologischen Einordnung des Delhi-Sultanates unter Muhammad Ibn Tugluq, IU, 165 (Berlin,

1993)

Cunningham, Sir Alexander, 'Report of a tour in Bundelkhand and Rewa in 1883-84', ASIR 21

(1885), 1-137

Dani, Ahmad Hasan, 'ShamsudDin Ilyas Shah, Shah-i Bangalah', in Gupta et ah (eds.), 50-8

Dardess, John W., 'From Mongol empire to Yuan dynasty: changing forms of imperial rule in

Mongolia and Central Asia', Monumenta Serica 30 (1972—3), 117-65

Day, U. N., The government of the Sultanate, 2nd edn (New Delhi, 1993)

'The north-west frontier of the Sultanate', in his Some aspects of medieval Indian history (New

Delhi, 1971), 29-57

'The north-west frontier under the Khalji Sultans of Delhi', IC 39 (1963), 98-108

De Bary, W. T. (ed.), Sources of Indian tradition (New York, 1958)

Derrett, J. Duncan M., The Hoysalas: a medieval Indian royal family (Oxford, 1957)

Deyell, John S., Living without silver: the monetary history of early medieval North India (Oxford and Delhi, 1990)

Digby, Simon, 'Iletmish or Iltutmish? A reconsideration of the name of the Dehli Sultan', Iran 8

(1970), 57-64

'The currency system', in Raychaudhuri and Habib, (eds.), Cambridge economic history, 93-101

'The sufi shaykh and the sultan: a conflict of claims to authority in medieval India', Iran 28 (1990),

71-81

'The sufi shaikh as a source of authority in mediaeval India', in Gaborieau (ed.), Islam et societe,

55-77

War-horse and elephant in the Delhi Sultanate: a problem of military supplies (Oxford and

Karachi, 1971)

Dikshit, R. K., The Candellas of Jejakabhukti (New Delhi, 1977)

Doerfer, Gerhard, Turkische und mongolische Elemente im Neupersischen (Wiesbaden, 1963-75, 4

vols.)

Eaton, Richard M., The rise of Islam and the Bengal frontier, 1204-1760 (Berkeley and Los Angeles, 1993)

Ernst, Carl W., Eternal garden. Mysticism, history, and politics at a South Asian sufi center

(Albany, New York, 1992)

Friedmann, Yohanan, 'A contribution to the early history of Islam in India', in M. Rosen-Ayalon

(ed.), Studies in memory of Gas ton Wiet (Jerusalem, 1977), 309-33

Friedmann, Yohanan (ed.), Islam in Asia, I. South Asia (Jerusalem, 1984)

Gaborieau, Marc (ed.), Islam et societe en Asie du Sud, Collection Purusarthe 9 (Paris, 1986)

Golden, P. B., 'Cumanica II. The Olberli (Olperli): the fortunes and misfortunes of an Inner Asian

nomadic clan', AEMA 6 (1986 [1988]), 5-29

Gopal, Lallanji, The economic life of northern India, c. A.D. 700-1200, 2nd edn (Delhi, 1989)

Gupta, Hari Ram, et al (eds.), Essays presented to Sir Jadunath Sarkar, Sir Jadunath Sarkar

commemoration volume, II (Hoshiarpur, 1958)

Gupta, Satya Prakash, 'Jhain of the Delhi Sultanate', MIM 3 (1975), 209-15

Habib, Irfan, 'Agrarian economy', in Raychaudhuri and Habib (eds.), Cambridge economic history, 48-76

An atlas of the Mughal empire (Delhi, 1982; repr. 1986)

'Barani's theory of the history of the Delhi Sultanate', IHR 7 (1980-1) 99-115 'Economic history of

the Delhi Sultanate - an essay in interpretation' IHR 4 (1977), 287-303

'Formation of the Sultanate ruling class of the thirteenth century', in I. Habib (ed.), Medieval India

1, 1-21

'Non-agricultural production and urban economy', in Raychaudhuri and Habib (eds.), Cambridge economic history, 76-93

'The price regulations of 'Ala'udDin Khalji - a defence of Zia' Barani', IESHR 21 (1984), 393-414

Habib, Irfan (ed.), Medieval India I. Researches in the history of India 1200-1750 (Oxford and

Delhi, 1992)

Habib, Mohammad, and Nizami, Khaliq Ahmad (eds.), The Delhi Sultanat (A.D. 1206-1526), A

comprehensive history of India, V (Delhi, 1970)

Habibullah, A. B. M., The foundation of Muslim rule in India, 2nd edn (Allahabad, 1961)

Haig, Sir Wolseley, 'Five questions in the history of the Tughluq dynasty of Dihli', JRAS (1922),

319-72

Haig, Sir Wolseley (ed.), The Cambridge history of India, III. Turks and Afghans (Cambridge,

1928)

Hambly, Gavin R. G., 'Twilight of Tughluqid Delhi', in R. E. Frykenberg (ed.), Delhi through the

ages: essays in urban history, culture and society (Oxford and Delhi, 1986), 47-56

'Who were the Chihilgani, the Forty Slaves of Sultan Shams al-Din Iltutmish of Delhi?', Iran 10 (1972), 57-62

Hardy, Peter, 'Didactic historical writing in Indian Islam: Ziya al-Din Baranl's treatment of the

reign of Sultan Muhammad Tughluq (1324-1351)', in Fried-mann (ed.), Islam in Asia, I, 38-59

'Dihli Sultanate', Enc.Isl2

'Djizya, hi. India', Enc.Isl2

'Force and violence in Indo-Persian writing on history and government in medieval South Asia', in

Milton Israel and N. K. Wagle (eds.), Islamic society and culture. Essays in honour of Professor Aziz Ahmad (Delhi, 1983), 165-208

Historians of medieval India (London, 1960)

'The authority of Muslim kings in mediaeval South Asia', in Gaborieau (ed.), Islam et societe, 37-

55

'The growth of authority over a conquered political elite: the early Delhi Sultanate as a possible

case study', in John S. Richards (ed.), Kingship and authority in South Asia (Madison, Wisconsin, 1978),

192-214

'The oratio recta of Baranl's Ta'nkh-i Firuz Shahi - fact or fiction?', BSOAS 20 (1957), 315-21

Hodivala, S. H., Studies in Indo-Muslim history (Bombay, 1939-57, 2 vols.)

Husain, A. Mahdi, The rise and fall of Muhammad bin Tughluq (London, 1938) Tughluq dynasty

(Calcutta, 1963)

Irwin, Robert, The Middle East in the middle ages: the early Mamluk Sultanate 1250-1382

(London and Sydney, 1986)

Islam, Riazul, 'Some aspects of the economy of northern South Asia during the fourteenth

century', JCA 11 (1988), no. 2, 5-39.

'The rise of the Sammas in Sind', IC 22 (1948), 359-82

Islam, Zafarul, 'Firuz Shah's attitude towards non-Muslims - a reappraisal', IC 64 (1990), part 4,

65-79

'The Fatawa Firuz Shahi as a source for the socio-economic history of the Sultanate period', IC 60

(1986), part 2, 97-117

Jackson, Peter, 'Jalal al-Din, the Mongols and the Khwarazmian conquest of the Panjab and Sind', Iran 28 (1990), 45-54

'Kutb al-Din Aybak', Enc.Isl2

'The dissolution of the Mongol empire', CAJ 22 (1978), 186-244

'The fall of the Ghurid dynasty', in Carole Hillenbrand (ed.), Festschrift for Professor Edmund

Bosworth (EDinburgh; forthcoming)

'The Mamluk institution in early Muslim India', JRAS (1990), 340-58

'The Mongols and India, 1221-1351', unpublished PhD thesis, Cambridge, 1977

'The Mongols and the Delhi Sultanate in the reign of Muhammad Tughluq (1325-1351)', CAJ 19

(1975), 118-57

Jackson, Peter, and Lockhart, Laurence (eds.), The Cambridge history of Iran, VI. The Timurid

and Safavid periods (Cambridge, 1986)

Jahn, Karl, 'Zum Problem der mongolischen Eroberungen in Indien (13.-14. Jahrhundert)', in

Akten des XXIV. internationalen Orientalisten-Kongresses Munchen... 1957 (Wiesbaden, 1959), 617-19

Joshi, P. M., and Husain, A. Mahdi, 'Khaljis and Tughluqs in the Deccan', in Sherwani and Joshi, History of medieval Deccan, I, 29-55

Kehrer, Kenneth C, 'The economic policies of Ala-ud-Din Khalji', Journal of the Panjab

University Historical Society 16 (1963), 55-66

Kempiners, Russell G., Jr, 'Vassaf's Tajziyat al-Amsar wa Tazjiyat al-Acsar as a source for the

history of the Chaghadayid khanate', JAH 22 (1988), 160-87

Khuhro, Hamida (ed.), Sind through the centuries (Oxford and Karachi, 1981)

Kolff, Dirk H. A., Naukar, Rajput and Sepoy: the ethnohistory of the military labour market in

Hindustan, 1450-1850 (Cambridge, 1990)

Lal, Kishori Saran, History of the Khaljis A.D. 1290-1320, 3rd edn (Delhi, 1980) Twilight of the

Sultanate, revised edn (New Delhi, 1980)

Lessing, F. D., A Mongolian-English dictionary (Berkeley and Los Angeles, 1960)

Lewis, Bernard, The Jews of Islam (Princeton, 1984)

Majumdar, R. C, Chaulukyas of Gujarat (Bombay, 1956) History of medieval Bengal (Calcutta,

1973)

Manz, Beatrice Forbes, The rise and rule of Tamerlane (Cambridge, 1989)

'The ulus Chaghatay before and after Temur's rise to power: the transformation from tribal

confederation to army of conquest', CAJ 21 (1983), 79-100

Mirza, M. Wahid, The life and works of Amir Khusrau (Calcutta, 1935)

Misra, S. C, The rise of Muslim power in Gujarat, 2nd edn (New Delhi, 1982)

Moosvi, Shireen, 'Numismatic evidence and the economic history of the Delhi Sultanate', in PIHC 50. Gorakhpur 1989 (Delhi, 1990), 207-18

Moreland, W. H., The agrarian system of Moslem India (Cambridge, 1929)

Morgan, D. O., The Mongols (Oxford, 1986) Morgan, D. O. (ed.), Medieval historical writing in

the Christian and Islamic worlds (London, 1982)

Nigam, S. B. P., Nobility under the Sultans of Delhi A.D. 1206-1398 (Delhi, 1968)

Niyogi, Roma, The history of the Gdhadavdla dynasty (Calcutta, 1959)

Nizami, K. A., 'Early Indo-Muslim mystics and their attitude towards the state', IC 22 (1948), 387-98; 23 (1949), 13-21, 162-70, 312-21; 24 (1950), 60-71

On history and historians in medieval India (New Delhi, 1983)

Some aspects of religion and politics in India in the thirteenth century (Aligarh, 1961)

Studies in medieval Indian history (Aligarh, 1956)

Supplement to Elliot and Dowson's History of India, II—III (Delhi, 1981)

'The impact of Ibn Taimiyya on South Asia', JIS 1 (1990), 120-49

Nizami, K. A. (ed.), Politics and society during the early medieval period. Collected works of

Professor Mohammad Habib (New Delhi, 1974, 2 vols.)

Nizamu'd-Din, M., Introduction to the Jawdmi'u'l-hikayat, GMS, ns, VIII (London, 1929)

Pal, Dharam, "Ala'-ud-Din Khilji's Mongol policy', IC 21 (1947), 255-63 "Ala'-ud-Din's price

control system', IC 18 (1944), 45-52

Parry, V. J., and Yapp, M. E. (eds.), War, technology and society in the Middle East (Oxford and

London, 1975)

Pelliot, Paul, Notes on Marco Polo (Paris, 1959-73, 3 vols. with continuous pagination) Notes sur

I'histoire de la Horde d'Or (Paris, 1950)

Pelliot, P., and Hambis, L. (eds.), Histoire des campagnes de Gengis Khan. Cheng- wou Ts'in

tcheng-lou (Leiden, 1951, vol. I only)

Petech, Luciano, Mediaeval Nepal (c. 750-1480), Serie Orientale Roma, X (Rome, 1958)

Prasad, Ishwari, A history of the Qaraunah Turks in India (Allahabad, 1936, vol. I only)

Qureshi, I. H., The administration of the Sultanate of Dehli, 4th edn (Karachi, 1958)

Rabie, Hassanein, The financial system of Egypt A.H. 564-741/A.D. 1169-1341 (Oxford and

London, 1972)

Rashid, Shaikh Abdur, 'Price control under AlaudDin Khilji', in Proceedings of the All-Pakistan

History Conference. First Session, held at Karachi ... 1951 (Karachi, [n.d.]), 203-10

Rasonyi, L., 'Les noms de personnes imperatifs chez les peuples turques', AOH 15 (1962), 233-43

Ray, H. C, The dynastic history of northern India (Calcutta, 1931-5, 2 vols. With continuous pagination)

Raychaudhuri, T., and Habib, I. (eds.), The Cambridge economic history of India, I. c. 1200- c.

1750 (Cambridge, 1982)

Richards, J. F. (ed.), Precious metals in the later medieval and early modern worlds (Durham,

North Carolina, 1984)

Roemer, Hans Robert, 'Timur in Iran', in Jackson and Lockhart, 42-97

Roy, N. B., 'The transfer of capital from Delhi to Daulatabad', JIH 20 (1941), 159-80

Saeed, Mian Muhammad, The Sharqi Sultanate of Jaunpur (Karachi, 1972)

Saran, P., 'Politics and personalities in the reign of Nasir-ud-Din Mahmud the Slave [sic]', in his

Studies in medieval Indian history (Delhi, 1952), 223-48

'The economic policy and price control of AlaudDin Khalji', Bharatiya Vidya 11 (1950), 195-215;

repr. in his Studies (above)

Sauvaget, Jean, 'Noms et surnoms de Mamelouks', JA 238 (1950), 31-58

Sharma, Dasharatha, Early Chauhan dynasties, 2nd edn (Delhi, 1975)

Sherwani, H. K., and Joshi, P.M. (eds.), History of medieval Deccan (1295-1724) (Hyderabad,

AP, 1973, 2 vols.)

Siddiqi, M. Y. Z., 'Arzdasht of Badr Hajib', MIM 2 (1972), 291-7

Siddiq[u]i, Iqtidar Husain, 'Historical information in the thirteenth century collections of Persian

poems', Studies in Islam 19 (1982), 47-76

'The nobility under the Khalji Sultans', IC 37 (1963), 52-66

'Fresh light on Diya' al-Din Barani: the doyen of the Indo-Persian historians of medieval India', IC 63 (1989), 69-84

'Social mobility in the Delhi Sultanate', in I. Habib (ed.), Medieval India 1, 22-48

'Sultan Muhammad bin Tughluq's foreign policy: a reappraisal', IC 62 (1988), part 4, 1-22

'The Afghans and their emergence in India as ruling elite during the Delhi Sultanate period', CAJ

26 (1982), 241-61

'The Qarlugh kingdom in north-western India during the thirteenth century', IC 54(1980), 75-91

Smail, R. C, CrusaDing warfare 1097-1193 (Cambridge, 1956; 2nd edn, 1995)

Spuler, Berthold, Die Goldene Horde: die Mongolen in Rubland 1223-1502, 2nd edn (Wiesbaden,

1965)

Die Mongolen in Iran: Politik, Verwaltung und Kultur der Ilchanzeit 1220-1350, 4th edn (Leiden,

1985)

Stapleton, H. E., 'Contributions to the history and ethnology of north-eastern India - IV. Bengal

chronology during the period of independent Muslim rule. Part I, 686-735 A.H. (1286-1334 A.D.)', JASB ns 18 (1922), 407-30

Troll, Christian W. (ed.) Muslim shrines in India (Oxford and Delhi, 1989)

Venkata Ramanayya (Venkataramanyya), N., 'The date of the rebellions of Tilang and Kampila

against Sultan Muhammad bin Tughlaq', Indian Culture 5 (1938-9), 135-46,261-9

The early Muslim expansion in South India (Madras, 1942)

Verma, H. C, Dynamics of urban life in pre-Mughal India (New Delhi, 1986)

Welch, Anthony, and Crane, Howard, 'The Tughluqs: master-builders of the Delhi Sultanate', Muqarnas 1 (1983), 123-66

Wink, Andre, Al-Hind: the making of the Indo-Islamic world, I. Early medieval India and the

expansion of Islam, 7th-11th centuries (Leiden, 1990)

Al-Hind: the making of the Indo-Islamic world, II. The Slave Kings and the Islamic conquest,

11th-13th centuries (Leiden, 1997)

Woods, John E., 'The rise of Timurid historiography', Journal of Near Eastern Studies 46, part 2

(1987), 81-108

Yule, Sir Henry, and Burnell, A. C, Hobson-Jobson: a glossary of colloquial Anglo-Indian words

and phrases, new edn by W. Crooke (London, 1903)

Index

Abachi (Arslan Khan), 85

Abachi (Mongol amir in India), 174

Abachi (Neguderi chief), 218, 225

cAbbasid Caliphate, at Baghdad, 3,4,37,44-5, 108; at Cairo, 162, 271-2, 296; diploma for Delhi

sovereigns from, 6,37-8,44-5,162-3, 167, 234, 272, 296, 298; diplomas sent to other Indian monarchs, 322;

Caliphs: al-Hakim, 272, 296; al-Mustakfi, 271-2; al-Mustansir, 37-8; al-Mu'tadid, 296, 298; al-Mutawakkil,

296, 298; al-Nasir, 6, 37

'Abd-Allah (Chaghadayid prince), 118, 121, 218

'Abd-Allah 'daftar-khwan', 261

cAbd-Allah Shah Changal, 211

Abu, Mount, 10, 12, 132

Abu Bakr Shah (sultan 1389-90), 190, 306-8,313,318

Abu Muslim (muhrdar), 185

Abu Rija, family, 181-2; Mujir al-Din, 181, 210, 213; Husam al-Din, 181-2; Shams al-Din,155n,

182, 305

Abuhar, 67n, 127-8, 249

Aden, 186

'Adilabad, 259

'Adud al-Din Yazdi, Sayyid, 184, 233n

Afghanistan, 7, 115, 185, 217-20, 224-6, 263, 313

Afghanpur (in Katehr), 243-4

Afghanpur (near Delhi), 161, 330

Afghans, 11, 62, 174, 184, 188, 274, 298, 323

'Afif, Shams-i Siraj, 21-2, 90, 155-6, 159, 167-70, 282; lost works of, 152; his family,

155,206,248-9,314

Africans, see Habashis

Agra, 324

Agroha, 266n

Ahmad b. *Tulabugha, 201

Ahmad-i Ayaz, see Khwaja Jahan

Ahmad-i Chap, 83, 200n

Ahmad-i *ChhItam ('Malik Qirabeg'), 176, 189, 195; his sons, 189n

Ahmad-i Iqbal, 185-6

Ajayagarh, 143, 215

Ajmer, 9, 10, 19-21, 130, 132

Ajudhan, 127-8

Akalkot, 212

Akola (Ankula), 213n

cAla' al-Din Ajudhani, Shaykh, 167

cAla' al-Din 'All Shah ('Ali Mubarak; sultan in Bengal), 267

'Ala'al-Din Ayaz, 189

'Ala' al-Din Muhammad (later Malik Chhajju), 54, 76-8, 82, 84, 101, 125, 139, 191,248

'Ala' al-Din Muhammad b. Tekish (Khwarazmshah), 11, 13, 32, 42n

'Ala' al-Din Muhammad Shah Khalji (sultan 1296-1316), xiii, 45, 50, 57, 59, 83, 87, 95, 123, 125,

133, 135, 138, 145, 159-60, 182, 190-1, 193, 208, 278n, 279, 291; lost sources on, 152; Baranis depiction

of, 156-7, 159-60, 175-6; his seizure of the throne, 54, 100,124; his reign, 156-7, 171-6; and the Mongols, 104, 173-4, 219, 221-4, 227-30; warfare against Hindu powers, 127, 146-7, 194-202; his nobles cAla'is),

171-80, 188-9, 191; his administrative and economic reforms, 156, 173,213,238-49,279

'Ala' al-Mulk (Barani's uncle), 172, 208

'Alam Shah, 'Ala' al-Din (Sayyid ruler 1445-51), 323

'Alapur, 126, 183, 199n

'All Beg (Mongol general), 174, 227-8, 230-1

'Ali Shah Kar, 271,273

'Ali Shah Kuhijudi, 'Ayn al-Din, 79, 84

'Ali-Wahan, 228

'Ali-yi Haydar, 179

'Ali-yi Isma'il, 26, 29

'Ali-yi Mardan, cAla' al-Din (sultan in Bengal), 28-30, 36

'Alim al-Mulk, Nizam al-Din, 181, 273, 275

Alp Khan (Muhammad), Nizam al-Mulk, 181

Alp Khan (Sanjar), 171, 176-7, 188, 196, 220, 236-7, 287-8

*Altunapa, Ikhtiyar al-Din, 67, 70, 97, 128 Alughu (Mongol amir), 83, 118, 178, 200n Alur, 205

Alwar, 128, 307-9, 322, 324; see also Meos cAmid Sunnami (poet), 45 amir-hdjib, 25

Amir Arslan *Kalahi, 51, 152

Amir Hasan Dihlawi, 51, 117, 154, 159

amir-i akhur, 25

amir-i dad, 25-6, 97

amir-i shikar, 25

Amir Khusraw Dihlawi, 49-51, 55, 78, 95, 116-17, 135, 153, 229, 235, 237; on

Hindus, 213, 290

amiran-i sada, 162, 183, 272-7, 303, 307

Amroha, 78-9, 100n, 136, 138, 176, 183, 227, 243-4, 269, 274, 308

Anatolia, 17,76,294

Anban (Mongol general), 105 Anbar, 112

Andkhud (Andkhoy), 11-13, 30n, 218n

Anhilwara, see Nahrwala Arabia, 214; Arabian Sea, 30, 252; Arabs, 41, 185

Arakan, 187

Aram Shah (ruler 1210), 29-30, 41-2

Aram Shah b. Khurram Kuhijudi, 189

Arangal, 161, 172, 180-2, 194, 199, 204-5, 209-11, 214-16; as Sultanpur, 205

Ardabili, cAbd al-cAziz, 163

Arghandab, 218 carid, 25

Arigh Boke (Mongol qaghan), 108-9, 116

Armenia, Lesser, 114

Army, of Delhi, 238-42, 247; numbers, 238-40, 260-1, 314; remuneration, 240-2, 261-2, 316-17;

decline of, 268-9, 314-18; see also duaspa, cavalry, murattab

Arran, 112

Arslan Khan, Taj al-Din Sanjar, 70, 74, 89-90,92-3,96, 101, 111

Asad al-Din Arslan, 179 Asawul, 274 Asi, 10

Assam, see Kamrup Assassins, 108 Awadh, 12, 19, 24, 28-9, 37, 42, 69, 74, 78, 82,84,86-7,90-

4,97, 100-1, 111n, 118, 135, 139, 144, 172-4, 191, 204, 251, 266, 270, 301, 307; town of, 138; Muslim

advance in, 138,200 'Awfi, 8, 31,41, 193

Awrangzib (Mughal emperor), 239

Ayazi dynasty, 89; see also Kabir Khan Ayaz; Taj al-Din Abu Bakr b.

Ayaz Ayba, Rukn al-Din, 178n

Aybeg, Qutb al-Din (ruler 1206-10), 7, 11-13, 19-21,24-5,42, 124, 134, 143, 145-6, 195; his early

career and status, 26-9, 31, 62; his nobles (Qutbis), 29, 42

Aybeg-i Tamghaj, 26

cAyn al-Mulk Husayn al-Ash'ari, 40

'Ayn al-Mulk Multani, 173, 177, 179-80; 199,202,214-15,329 cAyn al-Mulk, see Ibn Mahru

Aytegin, Ikhtiyar al-Din (na'ib), 57, 63n, 67

Aytegin-i Mui-yi Daraz, Ikhtiyar al-Din (later Amin Khan), 71,76, 78, 94

Aytemur *Kachhan, 81-2

Aytemur Surkha, 81-3

A'zam Malik Shaykhzada Bistami, 186

Azerbaijan, 33, 112

'Aziz Khammar, 185, 274

Babur, Zahir al-Din Muhammad (Timurid prince and first Mughal emperor), 18, 263, 322-5

Badakhshan, 105n, 115, 218n

Bada'un, 12, 13,26, 29,40, 69, 72, 74, 86n, 90, 99-101, 136, 138, 173, 187, 191, 243, 302, 323

Bada'uni, cAbd al-Qadir, 151

Badr (hajib), 229-30

Badr al-Din Sonqur-i Rumi, 63n, 67-8, 70, 101

Badr al-Din Sonqur, see Nusrat Khan Badr al-Habashi, 183

Badr-iChach, 152-3

Bagarkot, 212

Baghdad, 5, 37, 44, 70, 108-9

Bagh-i Jud (at Delhi), 29, 232n

Baghlan, 105n, 224

Baglana, 195-6, 275

Baha' al-Din Garshasp, 179, 181-2, 188, 199n, 203, 206, 211, 231. 256-7, 287

Baha' al-Din Toghril, 27, 63n, 98, 101, 143

Baha' al-Din Zakariyya, Shaykh, 323

Bahadur, Malik, 82n

Bahadur 'Bura', Ghiyath al-Din (sultan of Bengal), 200-1,257

Bahadur Nahir (Meo chief), 307, 309, 313

Bahalim, clan, 192

Bahlul Lodi (sultan 1451-89), 322-4

Bahmanids, dynasty, 162-3, 186, 211-12, 276, 294, 299; see also Hasan Gangu Bahraich, 31,73-4,

93, 139

Bahrain (malik of Ghazna), 185 Bahrain Khan (adopted son of Tughluq Shah I), 182,257,267

Bahram Shah b. lltutmish, Mu'izz al-Din (sultan 1240-2), 47, 55, 57-9, 91, 105, 128; his reign, 67-

8

Bahram-i Ayba, see Kushlu Khan Bakanawr, 298

Balaban, Ghiyath al-Din (sultan 1266-87; formerly Baha' al-Din and Ulugh Khan),

44,46,48,55,87,95, 100-1, 111, 113-14, 116-17, 123-5, 126n, 133-4, 141, 160, 171, 174, 189, 191, 238-9;

his origins, 63; career prior to his accession, 52, 62, 66n, 69-77, 98, 128, 129; operations against Hindu

powers, 55, 128-30, 132, 135-6, 138, 140, 142,144; his reign, 77-81,93-4; characterization of, by Barani,

51-2, 55, 253; his Turkish slaves, see Ghiyathis bala, bala-dast ('upper country'), 90, 179, 181, 263-4 al-

Baladhuri, 15,282

Balaram, 70, 134, 144, 309

Balga'un (Belgaum), 212

Balkh,217n,218n

Bamiyan, 6

Banaras, 10, 20, 29, 98, 138-9

Bang, 19, 92, 141-2; see also Bengal, Senas

Bangala, 19n

Bangarmaw, 200

Bar Ram, 118

baraka, 159-60, 170

Baran, 24, 26, 63, 70, 82, 96, 97, 106, 134, 223,244, 265

Barani, Diya' (al-Din), 21, 127, 159, 167; his ancestry, 50, 191, 292; relations with Muhammad b.

Tughluq, 164, 190; relations with Firuz Shah, 164, 167; death of, 155; his Fatawa-yi Jahandari, 21, 52, 164, 279, 291; his Ta'rikh-i Firuz-Shahi, 50-2, 54-5, 59, 65-6, 68-9, 103, 156-8, 162-8, 242, 246, 253-4,

279; variant recension of, 151, 155-6, 164, 167, 232, 243, 258, 270n, 330; his treatment of Muhammad's

reign, 162-6, 255, 292; his views on nobility of birth, 79, 189, 191, 292; attitudes towards Hindus, 279,

282, 290-2

bdrbeg, see amir-hajib

Barghund, 118n; Barghundi, Khudawandzada, 116

*Barihun (Barhamun?), 136

Barind (Varendra), 141-2

Baroda, 195,274,279

Barwala, 100

Basankot, 141

Bashir Mucizzi, Malik, 189

Bashir, Shaykh, 157

Basri (wazir), 5In, 53

Batu (Mongol khan), 107-8

Bayanchar, 81, 82n

baycat, 57-8

Bayhaqi, 59

Bayyu, Malik, 188

Beah, R., 113, 127,231

Begbars, Ikhtiyar al-Din, 78, 117, 238

Begtiit, Ikhtiyar al-Din, 84n

Bengal 9,17,24,28,49n,55,78,124-5,140, 161-2169,172,204n,215,257,261,298,302,314,Iltutmish

comquest, 36,7; Muslim advance in 13 19 140-2;

insubordination in 90-3; independent 13,19,140-2; Sultanate of (1287-1324) 53 94-5 142;

reduction of, by Tughluq Shah 200-1;

Berke(Mongal Ruler), 108-9,114,116n

Berki 29,42

Bernier, Francois, 239

Bhadga'un, 211

Bhakkar, 35, 80, 112 303

*Bhanawri, Abd al-Aziz-i[?] Shams, 152

Bharuch (Broach), 181, 196,273 275-6

Bhatigun, 139-40

Bhatner, 127-8, 167, 313

Bhattis, 127-8,248,280, 307

Bhayana, 71, 90, 96, 98, 101, 112 128, 134, 143-4,171,181,191,197,318,323-4

Bhilmal, 177

Bhilsan, 146

Bhiran, 186,271,273

Bhojpur, 135

Bhonga'un, 308

Bidar, 190, 205, 211,269, 273

Bihamadkhani, Muhammad, 151, 156, 169-70, 191, 318, 320, 322

Bihar, 9, 12-13, 18, 36, 91, 95n, 96, 125, 139-40, 188,307,309,323-4

Bijapur, 202n, 211,279

Bijapuri, 'Ayn al-Din, 50, 152

Bijnor, 136

Bilahur, 190

Bilram, 134n, 136

Binban, 36, 80, 105, 113-14, 116, 119; rulers of, see Ozbeg-bei; Hasan Qarluq; Nasir al-Din

Muhammad

Bini-yi Gaw, 115,218,224-5

Birdhul (Viradhavelan), 206, 208, 214

Birinjin, 83

Birjand (Barghund?), 118

al-Biruni, 14

Bistami, Jamal al-Din (shaykh al-islam), 101

Bodhan,205,211 Bojei(Chaghadayi prince) 121

Boroldai (Mongol amir), 2-7 Brahmans, 280, 286,288-9,291-3

Bhrmaputra R.,141-2

Bughra khan,Nasir al-Din Mahmud b.

Balaban,50,53-4,87,94-5,117-18,238

Bukhara, 63. 110. 116n

Bundi. 132

*Buququb,Malik78,99,101,189

Burhan al-Din(Alim Malik), 180-1

Burhan al-Din.Shaykh,159

Burma. 261

Bust, 11,220,225

Cairo, 272, 296, 322

Caliph, title of, adopted by Delhi Sultan, 158

caste system, 14

Caucasus, 109

Cavalry, heavily armoured (bar-gustuwan), 17-18, 90n, 125, 241; see also murattab

Ceylon (Sarandib), 193, 298

Chach-Nama, 15,283-4

Chaghadai (son of Chinggis Khan), 107

Chaghadayid khanate, 109-10, 121-2, 217-18, 226-7, 230, 233-6, 256, 263-4, 266, 311-12, 314;

khans: Alughu, 109-10, 116, 121; Baraq, 110, 119, 217, 235-6; Buzun, 233-4; Changshi, 233; Dore Temur,

233; Du'a, 217-21, 224, 227-8, 236; Esen Buqa, 224-6, 230; Kebek, 226; Khaffl, 234; Konchek, 224, 230;

Mahmud, 313; Tarmashirin, 164n, 227, 231 n, 232-4, 257, 262-4, 266, 313-14; Tughluq Temur, 311-12;

see also Mongols; Da'ud Qocha; It Qul; Qutlugh Qocha; Temiir Buqa; Yasa'ur

Chambal, R., 146

Chanchiwal, 211

Chandagarh (Chandrapur), 213

Chandawar, 10, 134, 136, 308

Chandellas, dynasty, 9, 10, 143-5, 199; kings: Hammiravarman, 199; Paramardideva, 12, 143;

Trailokyavarman, 143, 145n, 215; Viravarman, 143-4; Viravarman II, 199

Chanderi, 100,144-5, 174, 199, 323-4

Chanisar, Sinan al-Din (Sumra ruler), 35

Chapar (Mongol khan), 220, 224

chard'i (grazing-tax), 243, 248, 263

chatr, 28, 87

Chaulukyas, dynasty, 9, 10, 12,195; see also Gujarat

chawdhuri, 248-9

Chawhans, dynasty, 9, 12, 30, 128, 130, 132-3, 188, 323; see also Etawa; Jalor; Ranthanbor

Chawpala (Moradabad), 99, 138

Chedi, 143

Chenab, R., 113, 127,311

Chhajju, Malik, see 'Ala' al-Din Muhammad

Chihilganis, 65-6

China, 63, 108-10, 112, 184, 226, 252, 264, 288

Chinggis Khan, 13, 32, 34, 38, 103-4, 106, 312

Chishtiyya, sufi order, 159-61, 163

Chitor, 87, 132-3, 175, 215, 223, 253; reduction of, 197-8; as 'Khidrabad', 198; Guhila kings of:

Jaitrasimha, 133; Samarasimha, 133, 197; Ratan Singh, 197n

Chormaghun (Mongol general), 104

Chudasamas, 196

Chunar, 138-9

coinage, 37, 251, 261-2, 279, 316, 325

Crimea, 64

crossbow, see nawak

Dabhoi, 274

Dahhak, 163n

'Dalaklmalaki, 145

Dalmaw, 200, 265, 309

Damghani, Shams al-Din, 303

Damhai, 243, 244n

Damoh, 192n, 199

Damrila, 87, 126-7, 188

Dangiri, 211

Danuj Rai, 141

Daral-Harb, 6, 126,288, 301

Dar al-Islam, 6, 126

Dastiir Khan, Kamal al-Din b. cUmar, 306

Da'ud Khan b. Bayyu, 305n

Da'ud Qocha (Chaghadayid prince), 225-6

Dawar Malik, 166, 190

Dawlat Khan (ruler 1412-14), 321

Dawlat Khan Lodi, 324

Dawlatabad, 159, 162, 181, 186, 251, 270, 273, 275-6, 299; as second capital, 164-5, 210,232,256-

60

Daybul, 8, 21, 33, 35, 87, 106, 126

Dayir Noyan (Mongol general), 105-6, 108

Deccan, 124, 165, 169, 181, 183, 196, 199, 231, 251, 269-70, 272-6; reduction of, 158, 201-3;

Muslim colonization in, 210-12; loss of, 162, 275-6; as independent sultanate, 50, 163, 299; see also

Bahmanids; Dawlatabad; Deogir; Yadavas

Delhi, 9-10, 12-13, 19, 26; leaDing Muslim citizens of, 58-9, 83, 165; famine in, 244, 265-6;

exposure to Meo raids, 128; objective of Mongol invasions, 222-4, 236; fortification of, 231, 259-60; old

city of, 259-60, 310; transfer of population to Dawlatabad, 164-5, 232, 258-60; struggle for control of after

1388, 306-11; sacked by Temur, 313, 320

Delhi Sultanate: creation of, 13, 26, 29-38; designations of, in the sources, 86; a collection of sub-

kingdoms in the thirteenth century, 87-8; indirect nature of Muslim rule, 19, 124-6, 194; move towards direct rule, 253-5; extent of, 210; nobility of, 41-3, 61-85, 171-92; reasons for victories of, 213-16;

administration of, 95-102, 238-51 passim, 256, 262-3, 269-70, 273, 279, 284-7, 316-18; commerce of, 252-

3, 266; economic problems of, 255-6, 261, 265-6; disintegration of, 308-9, 318-20; Delhi Sultans, and

Islam, 278-9; Indian cultural influences on, 281; ruling dynasties of, see

Shamsids, Ghiyathids, Khaljis, Tughluqids; see also army, taxation Deogir, 87, 100, 146-7, 173,

175, 181, 191, 194-5, 196n, 201-3, 207, 209-11, 214, 254, 298; as Qutbabad, 203; see also

Dawlatabad, Deccan Deokot, 28, 141

Deoli, 135

Deopalpur, 77,99, 127-8, 157, 178, 182, 191, 223,243,310-11,313,319,323

Dhandh, R., 228n; Greater Dhandh, R., 117

Dhar, 179, 198-9, 251, 274, 276

Dhara, 186

Dharagir, 275-6

Dharmasvamin, 140

Dhimmis ('protected peoples'), 281-3, 295 dihliwal, 37

Dilan, Malik, 187 al-Dimishqi, Shams al-Din, 240

Dinar, Malik (later Zafar Khan), 175, 177

Diyarbakr, 112

Doab, 19, 86, 95, 231-2, 243-4, 252, 263, 265, 308, 317-18, 321-2; Muslim advance in, 134-6,200

Dorbei (Mongol general), 34, 106 Du'a, see Chaghadayids duaspa, 241 durbash, 30, 87

Dvarasamudra, 193, 202-3, 206, 208, 209n, 282; see also Hoysalas

Eastern Gangas, dynasty, 142, 206; see also Jajnagar Egypt, 45, 61-2, 66, 69-70, 80, 115, 223,

225, 252, 266, 294 elephants, war-, 90, 239, 306-7, 314, 324; in public ceremonies, 47, 281; gifts of, 93,

141; as plunder or tribute, 208-9, 302, 314; see also pilkhana Erach, 174, 199

Erkli Khan, 87, 118 Etawa, 10, 134-5, 169, 188, 302, 308-9, 323; see also Sumer, Uddharan

Europe, 110,252

Fakhr al-Din ('Fakhra'; sultan in Bengal), 182, 267 Fakhr al-Din, Malik al-Umara', 53, 58, 79, 83n,

84-5, 95, 172; his sons, 85

Fakhr al-Din'Ali Jawna, 172

Fakhr al-Din Hansawi, 180n

Fakhr al-Din *Qochu, 172

Fakhr-i Mudabbir, 8, 27, 281, 283-4

Farghana,18

Farid al-Din Ganj-shikar, Shaykh, 117, 168

Farrukhi, Fakhr al-Din Mubarak Shah, 68 Fars, 112, 119, 121, 193,219

Fath Khan, 169, 305, 332

fath-namas, 7, 152

Fatimids, 3

FirdawsT, 21

Firishta (Muhammad Qasim Hindu Shah Astarabadi), 50, 151

Firuz b. Taj al-Din Turk, Malikzada (later Firuz Khan), 188, 302-3, 306, 308

Firuz Khan b. Firuz Shah, 332

Firuz Shah b. Iltutmish, Rukn al-Din (sultan 1236), 40, 46, 55, 57, 59-60, 66, 76

Firuz Shah b. Rajab (sultan 1351-88), xiii, 21-2, 90, 102, 126, 155, 160, 185, 190, 194, 206, 229n, 240, 280, 288, 332; ancestry, 248-9; date of birth, 249n; early career, 182-3; his succession, 166-7; his

reign 167-70, 186-8, 296-306; as depicted in the sources, 167-70; and orthodoxy, 278, 286, 294-5; and his

Hindu subjects, 282, 286-9; and the nobility, 303-5; caliphal recognition of, 296, 298; promotes cultivation,

315; favour towards the religious classes, 168, 316; his military record, 169, 299-302; and the troops, 168,

317-18; and the Mongols, 311; and Temur, 312; his Futuhat-i Firuz-Shahi, 155, 168, 282; his slaves, 187-8,

306-8

Firuz Shah b. Zafar, 306

Firuz-i Aytegin, see Shams al-Din Firuz Shah

Firuzabad, 155, 305, 309-11, 321

Firuzkuh, 6, 21,25, 218n

Firuzpur, 302, 308, 318

furu-dast, 90

Gahadavalas, dynasty, 7, 9-10, 12, 18-19, 134-5, 138-9; kings: Jayachandra, 10, 19, 134;

Ajayasimha, 135

Gandhar, 183n, 196

Ganges, R., 6-7, 13, 54, 86, 91, 135, 138, 141, 228,309

Ganguri, 141

Ganjrut, 35

Ganuri,99,138, 243

Gardiz, 191

Garhmuktesar, 134

Gawr, 19, 125; see also Lakhnawti Gaya, 140

Ghaggar, R., 228, 313

Ghalib Khan, 306-7, 309-10

Gharchistan, 121,217-18 ghayr-wajhis, 316-17

Ghazi Malik, see Tughluq Shah I Ghazipur, 127

Ghazna (Ghaznayn), 5-6, 8, 10-13, 21 25, 28-30,32,34,36,39, 104-5, 112, 115, 118-19, 122, 180,

185,218-19.225,227, 229-31,236,312

Ghaznawi, Hajji cAbd al-Hamid. 154,282 284

Ghaznawids (Yaminids), 4-5, 6-7, 10, 16, 18, 24, 59, 62, 86, 192, 278, 280, 326; rulers: Bahram

Shah, 5, 7, 192; Ibrahim, 7; Khusraw Malik, 7-9, 21, 87; Khusraw Shah, 5; Mas'ud I, 5; Mas'ud III, 7;

Sebuktegin, 4; see also Mahmud of Ghazna

Ghiyath al-Din Mahmud Shah (sultan 1351), 166

Ghiyath al-Din Muhammad b. Iltutmish, 47, 69

Ghiyathids, dynasty, 44, 52-5, 58, 78, 82, 84, 171, 174, 189, 215, 333; in Bengal, 95, 141

Ghiyathis (slaves of Balaban), 76, 78, 82, 84 ghulam, see Turkish slaves

Ghur, 5, 7, 15, 28n, 34, 39-40, 112, 121, 217-18,236

Ghurids, 5, 6, 7, 11-12, 38, 86, 240, 278; conquests in India, 10-19; institutions of Ghurid empire, 24- 6; Delhi Sultans as heirs of, 31-2; sultans: 'Ala' al-Din Husayn Jahansuz, 5, 12; Ghiyath al-Din

Mahmud, 28; Ghiyath al-Din Muhammad b. Sam, 5-6, 9-10, 12, 21, 25n, 38n; see also Mu'izz al-Din

Muhammad b. Sam

Ghuris, 11,12,61,68-9

Ghuzz (Oghuz), 4-5, 11-12, 179, 326

Girnar (Junagarh), 196-7; 'Kanhgar', rana of, 197

Gogha(Quqa), 196

Golden Horde, 107-8, 110, 121, 226, 256, 312,314-15

Gondhiyana (Gondwana), 199

Gopalgir, 128

Gorakhpur, 139, 301

Govindaraja (Tomara king of Delhi), 10; his son, 19

grain supply, 244-5

Guhilas, 133; see also Chitor

Gujarat, 9, 45, 146, 161-2, 171, 173-5, 177, 178n, 181, 184, 187, 190-1, 196-7, 209, 215, 232n,

252, 272, 274-6, 298-301, 303-4, 307-11,317, 319; Muslim attacks on, 6, 10, 12, 19, 193, 195-6, 220n;

extent of conquest, 196-7; independent Sultanate of (from c. 1407), 182, 188, 196, 319; see also

Chaulukyas, Vaghelas

Gulbarga, 186,211,271,273

guruha, 214

Gutti, 203

Guyug (Mongol qaghan), 107

Gwaliyor, 10, 12, 20, 26-7, 38, 46, 96, 126, 134, 143-5, 157, 176, 188, 200, 321, 323-4

Habashis (black Africans), 61-2, 67, 68, 75, 183,252

Hajjaj (sultan of Kirman), 79-80

HajjiKe'un, 184

Hajji Mawla, 85, 241

Hajji Rajab Burqu'i, 272

Hajji Sa'id Sarsari, 272

Hamid al-Din (imam), 101

Hamid al-Din (nd'ib-i wakil-i dar to Sultan 'Ala' al-Din), 173,176

Hamid al-Din, Shaykh, 161

Hamid al-Din Multani, 175

Hamid Qalandar, 154

Hammiramahakavya, 123

Hanafi school, 242, 262, 282, 291

Hansi, 4, 7, 70, 72, 76, 98, 100-1, 128, 130, 159, 180,192,223,268

Harawi, Nizam al-Din Ahmad, 151

'Hardu Dal', 138

Hardwar (Bardar), 136, 313

Hariraja (Chawhan prince), 12, 130

Hariraja (feudatory of the Chandellas), 144

Hariyana, 128

'Harpal Deo', 202-3

Hasan (brother of Nasir al-Din Khusraw ' Shah), 177n

Hasan Gangu (Zafar Khan; 'Ala' al-Din ' Bahman Shah), 162, 212, 276, 298n, 299

Hasan-i Arnab, 26

Hasan-i Nizami, 7-8, 282

Hasan Qarluq, Sayf al-Din (Wafa Malik), 34, 36, 71, 88, 89, 104-6

Hashtnaghar, 230, 232

Hatya Paik, 79, 83 hawali (environs of Delhi), 95, 100

Hawd-i Khass, 222

Hawd-i Shamsi, 128

Haybat Khan, 101

Hazar Sutun, palace, 177, 260

Herat, 5-6, 21, 108, 111, 121, 186, 219, 225-6, 234-5, 312, 322; kings of: Mu'izz al-Din Husayn,

234-5; Shams al-Din Muhammad Kart, 111-12,116; Shams al-Din Muhammad II, 121

Hinawr, 298 Hindu (Mongol general), 231

Hindu armies, alleged inferiority of, 213-16

Hindu chiefs, 9-10, 21, 124, 242, 247-50, 262, 285, 301

Hindu Khan, 62, 88, 89n

Hindu Kush, 103,227,312

Hindu temples, destruction of, 20, 146, 168, 289; preservation or repair of, 20, 282, 287-8,301

Hindu troops in Muslim armies, 21, 125, 280, 285 Hindu-Muslim relations within the Sultanate,

281,289-94

Hindustan, 86, 88; in narrow sense (lands E. of the Yamuna), 19, 69, 86-7, 90, 125, 135, 187,197,302

Hiranmar,Ayn al-Din (later Amin Khan), 84n,85

Hisar Firuza, 315 holy war (jihad, ghaza), 6-7, 18-22, 98, 158, 168, 298; aims of, 123-4, 208-9,

253-4 Hooghly, R., 142 horse archers, 17, 214 horses, as plunder, 208-9, 237; warhorses, 15, 264, 315;

possible dearth of, in Hindu armies, 214-15

Hoysalas, 193, 206; king: Ballala III, 206, 213n, 257, 293; see also Dvarasamudra

Hukayri (Hakeri), 212

Hulechu (Mongol chief), 268

Hiilegu, see Ilkhanate Humayun (Mughal emperor), 324 hunt, 25, 187; as military training, 240;

see also nerge Hurmuz, 219

Husam al-Din (Barani's grandfather), 50

Husam al-Din Oghulbeg, 26, 42, 98, 138

Husam al-Mulk (Husam al-Din b. Nuwa), ' 186

Husam Khan, 152-3

Husayn (Mongol amir), 311-12

Husayn b. Ahmad-i Iqbal, 186

Husdurg, 203 Hushang, Malik Taj al-Din, 177, 179, 204, 268

Ibn Abi'l-Fada'il, al-Mufaddal, 154-5

Ibn al-Athir, 8, 13,33

Ibn Battuta, 17, 51, 128, 151, 155, 162-4, 178, 183n, 184-5, 210, 227, 229, 234, 265, 293 Ibn

Hassul, 64 Ibn Mahru, cAyn al-Mulk, 125-7, 153-4, 180-2, 185, 270-1, 274, 282, 286, 303, 305n,

316-17, 329 Ibn al-Sa'i, 18n, 44 Ibn Sa'id al-Maghribi, 239 Ibn Taymiyya, 163 Ibrahim (sultan 1517-26),

324 Ibrahim b. Hasan, Sayyid, 192 Ibrahim, Rukn al-Din (sultan 1296), 45n, 55 Iftikhar al-Din

Muhammad-i cUmar, 41 Iftikhar al-Mulk Sharaf al-Din Muhammad Rashidi, 77n ijtihad, 163 Ikdala, 299

Ikhtisan-i Dabir, 153, 201, 233n Ikhtiyar al-Din 'Ali b. Aybeg (later Khan Jahan; 'Hatim Khan'),

78, 82n Ikhtiyar al-Din Alp Ghazi, 78, 82n Ikhtiyar al-Din Dawlat Shah (Bilge Malik), 37

Ikhtiyar al-Din Ghazi Shah (sultan in Bengal), 267

Ikhtiyar al-Din-i Hindu Khan-i Ghiyathi, 84

Ikit Khan (Sulayman Shah), 171, 173-5, 280 illel ('peace', 'submission'), 104, 121

Ilkhanate, 109-10, 114, 119, 219-20 225-6 233, 235-6, 256, 264, 312;

Ilkhans; Abaqa, 80n, 119, 121; Abu Said, 226, 233; Arghun, 119, 121;Ghazan, 49, 121,219-

20,236; Hiilegii, 108, 112, 114-16, 237, 239-40; Musa, 184; Oljeitii (formerly Kharbanda) 104, 219-20,

224-6; Togha Temur, 235

Iltutmish, Shams al-Din (sultan 1210-36), 8 13, 24, 26, 57, 62, 65, 84, 87, 96-7, 124, 126, 145,

160, 290-1; his reign, 29-43; suppression of his Muslim rivals, 34-41, 112-13; warfare with Hindu princes,

30, 39, 130, 132, 134-6, 146; and the Mongols, 33-4, 104-5; caliphal recognition of, 37-8; his death, 36,44,

55; succession to, 46; his slaves, see Shamsis

Ilyas Shah, Shams al-Din (sultan of Bengal), 267-8, 296, 299, 302

cImad al-Mulk, Malik Bashir, 187, 304, 306, 317

'Imad al-Mulk, Malik Shahin Sultani, 307

'Imad al-Mulk, Muhammad (carid), 11

'Imad al-Mulk, Sharaf al-Dawla Abu Bakr, 43 immigration, Muslim, into India, 39-43, 79-80, 278

ir cam (exempt from service), 101

India, wealth of, 252

Indian Muslims, 72-3, 175, 185, 188

Indian slaves, 79, 175, 177, 187; see also Firuz Shah b. Rajab, his slaves

Indians, in service of Delhi Sultans, 185-6, 188,271,273,279-80,294

Indrapat, 10, 232

Indri, 232

Indus, R., 3, 30, 32, 34-5, 90, 106, 126, 225, 228,233,251,276,313 informers (bandan, munhiyan),

100, 167, 245-6, 250, 256 Iqbal (Mongol general), 228, 230, 328 iqtac, 24, 28,95-8,241,249,269,304,316

Iraq112,1l4, 263

cIsami, 21, 50-1, 156, 162-5, 202, 250, 287, 292

Isfahan, 4

Ishaq b. cImad al-Mulk Bashir, 304, 317

Islam Khan, Mubashshir-i Chap, 307

Ismacil *Mukh, Nasir al-Din, 163, 274-6

Ismailis, 3-4, 7-8, 13, 108, 278n

It Qul (Chaghadayid prince), 225

cIwad, Ghiyath al-Din (formerly Husam al-Din; sultan of Bengal), 28, 36-7, 39, 139, 141,160

cIwad, Malik (Khalaj amir), 83

cIzz al-Din (brother of Nusrat Khan), 172

cIzz al-Din (dabir-i mamdlik to Sultan cAla' al-Din), 173, 176

'Izz al-Din cAlI, 27, 42 cIzz al-Din Bakhtiyar, 41

'Izz al-Din Balaban-i Yiizbegi, 72, 74, 92-3

'Izz al-Din *Bura Khan, 197

Jabalpur, 199

Jahanpanah, 259-60

Jahiz, 64

Jains, 280, 287-8

Jajapellas, dynasty, 143-5; king:

Chahadadeva ('Chahar-i Ajarl'), 144-5, 215

Jajmaw, 200

Jajnagar (Orissa), 21-2, 91-2, 94n, 142, 146, 155n, 169, 205-6, 209, 214-15, 299, 301-2, 314;

kings of: Narasimha II, 142; Virabhanudeva III, 206, 301; see also Eastern Gangas; term 'Jajnagar' used of Tipperah also, 141

Jajner, 111, 113, 191; Jajner river (the Sutlej?), 33

Jalal al-Din Ahsan Shah (sultan of Ma'bar), 192,267

Jalal al-Din Bukhari, Shaykh, 300

Jalal al-Din Firuz Shah KhaljT (sultan 1290-6), 44, 49, 50, 55, 58, 80, 87, 100, 118, 125, 129, 169,

178, 285, 295; his background, 80, 116; activity prior to his accession, 82, 98, 118, 127-8; his reign, 53-5,

82-4; warfare against Hindus, 54-5, 127-8, 130, 132-3, 138, 139, 145-6; characterization of, by Barani, 54-5; his sons, 56, 85, 190, 219; his nobles, 83-5, 190

Jalal al-Din Mas'ud b. Iltutmish, 73-4, 89, 111-13, 116, 133n

Jalal al-Din *Mingbarni, 32-6, 38-9, 104

Jalali, 135,293

Jalesar, 134, 306, 309

Jalindhar, 221n

Jalor, 9, 30, 41, 130, 177, 178n, 179, 198, 253, 308; reduction of, 198; Chawhan kings of:

Chachigadeva, 130; Udayasimha, 130; Kanhadadeva, 198

Jam (in Khurasan), 6, 155

jam (title of Samma ruler in lower Sind), 169, 188, 300; see also Sammas

Jamaji, Baha' al-Din 'Ali b. Ahmad, 30-1

Jamal al-Din 'AIT (Amir Jamal) Khaljl, 80, 190

Jamal al-DTn Nlshapuri, 79

Jamkhandl, 212

Jammu, 21, 118, 129,313

Janani, 127

Jam, 'Ala' al-Din, 42, 68, 87, 101-2, 327

JaralT, 135

Jaran Manjur, 221

'Jaspal Sihra' (ruler of Kuh-i Jud), 129

Jats, 15, 127, 178,283

Ja'urchi, 81

Java, 298

Jawnpur (formerly Zafarabad), 309, 314, 318-19; as an independent state, 319, 321-4; reconquest

by the Lodls, 323 Jazlra, 112 Jejakabhukti (Bundelkhand), 9, 143; reduction of, 199 Jhajhar, 310 Jhayin,

132-3, 171, 197-8, 243-4; as Shahr-I Naw, 197

Jhelam, R., 36, 113, 127,227

Jhinjhana, 134 jihad, see holy war Jind, 100

jizya, 282-8, 291

Jochi (son of Chinggis Khan), 107; Jochids, his descendants, 108-9, 115, 315

Junaydi, Husam al-Din, 190, 315; Nizam al-Mulk (wazir; formerly Mu'ayyad al-Mulk), 35, 43, 46,

68, 190, 290; Rukn al-Din, 190, 306

Jurwas, 6

Juwayni, 33, 49, 103

Juzjani, 7, 18, 25, 38, 39-41, 47, 48-9, 56, 72, 91n, 99, 101, 144, 191, 281-2, 289-90; his Tabaqat-i

Nasiri, 7, 45-9, 51, 59, 88, 93, 103; his Nasiri-Nama, 145

Kabar, 138,200,243-4

Kabir al-Din (qadi), 101

Kabir al-Din b. Taj al-Din 'Iraqi, 101, 152

Kabir Khan Ayaz, 'Izz al-Din, 62n, 63n, 67-8,70,88-9,97,113

Kabul, 105, 112, 177, 227, 312-13, 322-3

Kafur, Malik (muhrdar), 179-80, 201 Kafur 'HazarDinari, Malik (Malik Na'ib), 153, 157, 171, 173-7, 194, 199, 201-4, 206-9, 211, 213-14, 227n, 228, 247, 280

Kailas, 205

Kaithal, 118, 127-8, 191-2, 223, 292 *Kaithun, 139n Kakatiyas, dynasty, 161,211;see also

Arangal; Rudradeva II; Tilang Kalachuris, 143

Kalanawr, 231 -2

Kalinjar, 9, 12, 143, 145

Kalpi, 156, 188, 191, 303, 306, 308, 318; as an independent principality, 320

Kalyani, 203, 210-11,287

Kama, 96, 128

Kamal al-Din, Hadr-i Jahan, 180

Kamal al-Din 'Gurg', 177, 198, 268

Kamal al-Din Ma'in, 309

Kamal-i Mahyar, 79

Kampil, 135

Kampila, 185, 203, 210, 257, 268

Kamrup (Assam), 13, 37, 92, 141-2

Kanauj, see Qinnawj

Kanbhaya (Cambay), 175, 193, 195-6, 252, 274-6

Kara,41,71,82,86,90-2,100,125,140,145-6, 172, 191, 197, 204, 269, 280, 299, 304

Karim al-Din Hamza, 42n

Karimi merchants, 252 karkhanas (imperial manufactories), 247 karwa, 11

Kasani, Jalal al-Din (grand qadi, c.1242), 68;

Jalal (grand qadi c.1292), 83, 101

Kashmir, 6, 36, 105, 115, 129, 237, 264, 298

Kasili, 130

Kasmandi, 138

Kasrak, 82n, 97, 136,243

Katehr (Rohilkhand), 55, 136, 138, 169, 200, 240,243, 266, 302, 323

Kathiawad, 195-6, 276

Kawlam, 298

kayasthas, 186

Kaykhusraw b. Muhammad b. Balaban, 53, 87,219 Kayqubad, Mucizz al-Din (sultan 1287-90), 44,

49-50, 55, 58, 60, 78, 94, 118, 126, 146, 157, 171, 174, 189, 219, 230; his ancestry, 57; his reign, 53-4, 81-

2

Kayumarth, Shams al-Din (sultan 1290), 44, 53,55,81

Keder (Mongol general), 221, 224, 237 Kerei, 81

*Kezlik Khan, 40, 96-8

Khalaj, 11-13, 18-19,27,44,53,61,80, 82-4,87, 140,171,273

khalisa, 95, 99, 138, 241, 244, 249, 304

Khaljis, dynasty, 44, 50, 54, 82, 157, 179, 191 and passim, 334;

'Khalji revolution', 82-5

Khamush, Malik, 83, 118n, 157

Khan-i Khanan, Mahmud, 83

Khanbaligh (Ta-tu), 109 10

Khandesh, 318 Khan Jahan (I), wazir (formerly Kannu; Qiwam al-Mulk Malik Maqbul), 169, 185-

6, 268, 275, 299, 303-4, 306

Khan Jahan (II), wazir, 305 Mara; (land-tax), 242-4, 250-1, 262, 283-4, 315; confusion with jizya,

284-5

Kharbanda (Oljeitii), see Ilkhans kharitadar, 192, 256

Kharonsa, 139n, 301

Khattab, Malik, 188 Khidr Khan (son of Sultan 'Ala' al-Din), 50, 87, 157-8, 160, 176, 198, 229-30,

293

Khidr Khan b. Sulayman (Sayyid ruler 1414-21), 304, 310, 318-19, 321-3; subordinate to Temur,

318- 19

Khitan, 62, 65

Khokhars, 13, 21, 62n, 127, 179, 237, 309, 323; chiefs: Gul Chand, 179, 268; *Samaj Rai, 179;

Shaikha, 309-10, 313

Khor, 243, 244n

Khudawandzada (daughter of Tughluq Shah I), 166

Khurasan, 5-6, 11, 13, 39,41, 80n, 110, 119, 121, 159, 178-9, 184, 219-20, 224-6, 229, 234; in

broader sense (lands W. of the Indus), 263-4;

'Khurasanls', 184-5, 271

Khurram Kuhijudi, 'Izz al-Din or Ikhtiyar al-Din, 79, 84, 189

khushddshiyya, 69

Khusraw Khan, see Khusraw Shah, Nasir al-Din

Khusraw Malik, 166n

Khusraw Shah, Nasir al-Din (sultan 1320; formerly Khusraw Khan; Hasan), 157-60, 177-80, 182,

202, 205, 207-9, 280

khut (khot), 124; see also Hindu chiefs

khutba, 4

Khuzistan, 112

khwaja (accountant), 100

Khwaja Hajji, Nasir al-Mulk Siraj al-Din, 173,177,180

Khwaja Jahan, Ahmad-i Ayaz (wazir), 162, 166-8, 183, 185-6, 189, 257, 268, 288n, 298

Khwaja Jahan, Malik Sarwar SultanI (Sultan al-Sharq), 307, 309-10, 314, 318-19

Khwaja Khatlr al-Din (wazir), 84

Khwarazm, 6, 11,42, 104

Khwarazmians, Khwarazmshahs, 6, 11, 20, 30, 32, 34, 70n, 326; invasion of India, 32-3, 40

Khyber Pass, 10

*Kiki, Malik, 84n

Kili, 222, 239

KTlokhrl, 53, 58, 70, 259

*Kirit Khan, Taj al-Din Sanjar, 63n, 68n, 91, 138

Kirman, 79, 112, 115, 119, 121,219,225

Kirmani, Muhammad b. Mubarak (Amir Khwurd), 154, 159

Kishli Khan, Sayf al-Din Aybeg, 63n, 64, 70-2, 74, 76, 80, 101, 134, 145n

Kochii, 77

'Koka Pradhan', 198,215

K61, 12, 20, 26, 68n, 84n, 96, 98, 134, 144, 223,244, 265

Konkan, 204, 268

Kopek (Mongol general), 228, 230-1, 328

Kotgir, 213

Kotla (Kopila?), 243, 244n

kotwal, 53, 96, 172, 180, 189

Koyir (Koher), 205, 211,273

KuchI, Malik Fakhr al-Din, 84; Malik Taj al-Din, 84; Malik NasTr al-Din, 84n

Kuh-i Jud, 13, 55, 79, 11 In, 113, 129

kuhpdya, 128; see also Alwar; Meos

Kuhram, 20, 24, 26, 30, 35, 77, 228

Kujah, 36, 111,113

Kumta (Kunti?), 203

Kunwarl (Kunar), R., 146

Kiirbiiz, 106

kuregen, 227n

Kiiresbe (Mongol prince), 218, 224

Kurraman, 36, 105, 118-19

Kiishlii Khan (Bahram-i Ayba), 174, 178, 182, 184-5,232,256-8,260 Kiishlii Khan, cIzz al-Din

Balaban, 48, 56, 63n, 69-72, 74-6, 89, 94, 101, 106, 114-16, 130; submits to Mongols, 112

Lahari, 251

Lahore (Lahawr), 5, 11, 13,21,26-8,30-1, 33-5, 42, 55, 67, 74, 77, 87-90, 92-4, 99, 104, 116, 178,

229, 231-2, 243, 268, 309-10, 322-4; a mulk, 87, 97; Mongol sack of (1241), 57, 105; subject to Mongols,

73,89, 111, 113

*LakchIr (Neguderi chief), 225

Lakhnaw (Lucknow), 138, 200

Lakhnawtl (formerly Gawr), 18-19, 37, 42, 47, 50, 53-4, 78, 88, 90-5, 99, 101-2, 117, 140-2, 153,

182, 189, 200, 257, 326; a mulk or iqlim, 87; known as 'Bulghakpur', 90

Lakhnor, 91, 141-2

Laksmanasena, 13, 289

Lawa, 130

Lodis, dynasty, 289, 321, 323-5

Lohrawat, 222

Ma'bar, 161-2, 174, 181, 183n, 194, 202, 204-5, 208-11, 213, 293, 298, see also Pandyas;

campaigns against, 206-7; revolt of, 258, 267-8; independent Sultanate of, 192, 207, 267-8, 298

MaDin, 6, 40n

Madura, 192, 207, 209, 267

maghribi (mangonel), 215

Mahandari, 212

Maha'un, 134, 143-4

Mahmud of Ghazna, 3-6, 8, 195, 214n, 291, 324; as model for Delhi Sultans, 3, 21-2

Mahmud, Malik, 84n

Mahmud b. Firuz Khan (later Nasir al-Din Mahmud Shah, ruler of Kalpi), 308, 318, 320

Mahmud Shah b. Iltutmish, Nasir al-Din (sultan 1246-66), 31, 44-8, 55-6, 58, 89, 98, 111-14, 123-

4, 134, 136, 139, 169, 290; his reign, 71-6; his death, 52, 76

Mahmud Shah II, Nasir al-Din (sultan 1396-1412), 309-11, 313-14, 317-21

Mahoba, 199,304,318

mahrusa ('reserved'), 96-7

Mains, 127-8, 307; see also Kamal al-Din

Makhduma-yi Jahan, 258

Makran, 7, 33, 112

Malabar, 193, 204

Maldives, 204

Malik Na'ib, see Kafur 'HazarDinari

malikal-tujjar, 181, 252

Mall, Malik Ikhtiyar al-Din, 174, 180n, 200, 275

Malta (later Iqbal Khan), 308n, 309-11, 313-14,318-19,321,323

Malwa, 6, 92, 133, 135, 147, 177, 179-80, 185, 194, 211, 213, 253, 265, 319; raided, 19, 144, 146;

reduction of, 198-9; as an independent state, 319-20, 322-3; Dilawar Khan ('Amid Shah), ruler of, 318-19;

Hushang Shah, sultan of, 320, 322

mamluk, see Turkish slaves

Mamluk Sultanate, in Egypt and Syria, xiii, 45, 62, 80, 108-10, 115, 154, 178, 236, 239, 269;

sultans: Baybars, 64; al-Nasir Muhammad b. Qala'un, 256, 261

Mandahars, 127-8

Mandalkhur, 199

Mandhol (Mudhol), 212

Mandiyana, 138

Mandor, 30, 39-40, 70, 130

Mandu, 199,215

Maner, 12, 139

Manglawr, 134

Manikpur, 86n, 91, 145

Mansura, 3-4

Mansurpur, 74n

Maqbul, Malik, see Khan Jahan (I)

Maram, 212

Marlgala, 101

Ma'ruf, Sayyid al-Hujjab, 190

Marwar, 324

Mas'ud Khan (son of Tughluq Shah I), 182

Mas'ud Shah b. Firuz Shah, cAla' al-Din (sultan 1242-6), 47, 55, 58-60, 69, 71, 91-2,97-8, 106; his

reign, 68

Mathura (Muttra; Mahir), 134, 143-4, 200

mawas[at], 99, 124-5 and passim

mawlazadas, 78

al-Mawsil, see Mosul

Mayapur, 134

Mengu Khan (nephew of Sultan cAla' al-Din), 173

Meos (Miwat), 55, 114, 128-9, 305, 307, 321, 323; see also Alwar, Bahadur Nahir

Mewar, 133,324

Miraj, 212

Mirat (Meerut), 12, 20-1, 74, 134, 232, 244, 313

Mlecchas, 123

Monggedii (Mongol general), 105-6

Mongke (Mongol qaghan), 75n, 106n, 107-9, 111,112n,119

Mongolia, 34, 103, 107, 184, 217

Mongols, 5, 13, 26, 49, 53, 54, 72, 73, 75, 79, 126, 129, 159, 172, 173, 175, 179, 182, 187n, 216,

238, 253, 290, 326; ideology of world domination, 103-4; conquest of Western Asia, 32, 38, 45, 64, 76,

112; appointment of residents (shihnas), 103, 112; first invasion of India, 33-4, 36n, 39, 104; subsequent

attacks on the Sultanate, 55, 57, 88-9, 94, 105-19, 219-32, 300, 311, 322; annual raids by, 106, 117;

numbers, 223, 228-9; purpose of invasions, 235-7; prisoners, 230-1, 236; women and children, 236;

embassies from, 104-5, 114, 144n, 184, 225, 239, 288; as harbingers of the last things, 113; and Indian

climate, 106; strife among, 93, 106-10, 115-16,219, 224-5, 229, 233, 311-12; counter-offensive against,

229-31, 263-4; friendly relations wi,th, 233-5; immigrants into India, 80-2, 106, 118, 174, 178, 185-6, 229,

234, 312; converts to Islam ('neo-Muslims'), 80-2, 172-4, 195, 197, 234; see also Chaghadayids; Chinggis

Khan; Golden Horde; Ilkhanate; Qaidu

Mosul, 8, 112

Motupalli, 205

Mu'ayyad al-Mulk Sajzi, 25

Mubarak, Malik, 186

Mubarak b. Qabul 'Khalifati', Malik, 84n

Mubarak Khan (son of Tughluq Shah I), 182

Mubarak Shah (Chaghadayid prince), 119; his sons, 121

Mubarak Shah (Sayyid ruler 1421-34), 322

Mufarrij Sultani, Malik Ikhtiyar al-Din (Farhat al-Mulk; Rasti Khan), 187, 307-8

Mughaltai, 81, 84n

Mughaltai (amir of Multan), 178

Muhadhdhab al-Din, Nizam al-Mulk (wazir), 67-8,71,96, 190

Muhammad b. Bakhtiyar, 12-13, 18-19, 24-6,28,39, 138-9, 141,326

Muhammad b. Balaban, Khan-i Shahid ('the Martyr Prince'), 53, 80, 87, 94, 100, 117, 126-7, 174,

237-8, 290, 328

Muhammad b. Qasim al-Thaqafi, 3, 6, 15, 282-3

Muhammad-i Shiran, 28

Muhammad Mulai (Shir Khan), 177

Muhammad Shah (muqta' of Tughluqpur), 188

Muhammad Shah (Sayyid ruler 1434-45), 322

Muhammad Shah b. Firuz, Nasir al-Din (sultan 1392-6; formerly Muhammad Khan), 302, 305-9,

312-13, 318, 332; lost manaqib of, by cAfif, 152; his slaves, 309

Muhammed shah b. Tughluq (sultan 1324-51), xiii, 124-5 128-9 139 174 189-92,197,208-9,213-

14,251,278,288, 292; early career (as Malik Jawna and UlughKhan), 179,180,181, 183,213; date of his accession, 330-1; his reign, 162, 182-6255-7image of, in the sources, 162-6, 255; his conquests, 197 199

203 205-7; his dealings with the Mongols, 231 -5; Khurasan' project, 240,260,262-4; issue of

'token'currency, 164-5 261-2-taxation, 262-3; revolts against,' 256-7,' 265-71,273-7; caliphal recognition of,

162, 271-2; his relations with sufi shaykhs, 160-1; his favour towards foreigners 41, 184-5, 233-4, 271,

294; his patronage of Indians, 185-6,271,280; and Hindu practices, 163, 287; his 'autobiography', 153; lost

manaqib of, by'Afif, 152

Muhammad Shah Lur, 178, 179n

muhtasib, 25, 245

Mu'in al-Din, Chishti shaykh, 159

Mu'in al-Mulk, Sama' al-Din b. cUmar, 306

Mu'izz al-Din Ajudhani, Suhrawardi shaykh, 161

Mu'izz al-Din 'Ali-yi cIwad, 36

Mu'izz al-Din Muhammad b. Sam (Ghurid sultan), 3, 5-8, 10-12, 16, 24-6, 43, 63n, 124, 129, 135,

324; his nobles (Mu'izzis), 12, 29, 42 mukus, see taxation, uncanonical

Multan, 3-4, 7-8, 13, 26-7, 30, 34-5, 39, 49, 71-2, 75, 81-2, 85, 87, 89, 94, 97, 104, 106, 111-13, 116-18, 126, 159, 168, 171, 174, 178, 183, 185, 219, 223, 228-9, 268, 270-1, 273-4, 303-4, 307, 310-11,

313, 316, 318-19, 329; independent, 323

Multanis, merchants, 241, 245-6, 250, 279

muqaddam, 124; see also Hindu chiefs

Muqarrab Khan, 309-10

Muqbil, Malik, 185n, 274

muqta', duties of, 97-100, 249-51, 269, 304; see also iqta'

murattab, 241

Musawi, Taj al-Din (mushrif), 68

Muslims, outside Delhi Sultanate, 193-4, 211, 293-4; deviant, 278; apostasy among, 293-4

Mutahhar, 152, 155

nadum ('boon-companion'), 51, 164, 178, 190

Nadol, 9

Nagada, 133

Nagarkot, 152n, 169, 194, 261, 282, 289, 301-2,306,311

Nagawr, 27, 42, 71-2, 130, 132, 161, 192, 228, 243,308,324

Nahrwala (Anhilwara), 9, 19, 33, 195-6, 275 na'ib (viceroy), office of, 67, 71-2, 77, 171-2, 175;

(deputy of a muqta'), 100

Nanadeva, chief of Salher and Mulher, 196, 275

Nanak, Malik, 175, 204n, 227, 280,

Nandana, 30, 34, 36, 104, 113

Nara'ina, 228, 236

NaranguT, 141

Narayan, 212, 294

Narbada, R., 194, 209

Narnawl, 128-9

Narwar, 143-5, 323-4

Nasawi, 33

Nasir al-Din (ruler of Sistan), 220

Nasir al-Din (sultan of Bengal), 200, 257

Nasir al-Din Abu Bakr b. Sun (Ghurid malik), 40

Nasir al-Din Aytemur (muqta' of Uchch), 27, 30n

Nasir al-Din Aytemur al-Baha'i, 35, 63n, 130, 132

Nasir al-Din Mahmud b. Iltutmish (d. 1229), 26,30,36-7,46-7,87, 138

Nasir al-Din Mahmud Chiragh-i Dihli, Chishti shaykh, 154, 161

Nasir al-Din Mardan Shah, 41 Nasir al-Din Muhammad b. Hasan Qarluq, 114,116

Nasir al-Din Muhammad b. Kushlu Khan, 116

Nasir al-Din Rana, 84n

Nasir al-Mulk, Malik Mardan Dawlat, 304, 311

nawak (crossbow), 16, 214

nawbat, 27, 38

Nawruz (Mongol general), 217-18

Nawruz Kuregen (Mongol chief), 234

Nawsari, 181, 196 nazir, 246

Negtider (Mongol general), 105, 115, 119;

Neguderis, 94, 115, 117-22, 178, 217-20, 224-6, 328; see also Qara'unas 'neo-Muslims', see Mongols, converts to Islam

Nepal, 139,201,302 nerge, 240

Nizam al-Din (dadbeg), 53, 79, 81, 84, 118, 126

Nizam al-Din (prince of Qays), 184 Nizam al-Din Awliya', Chishti shaykh, 154, 157, 159-61, 165,

167

Nizam al-Mulk (Seljukid wazir), 64

Nizam Ma'in, 269-70

Nudiya, 13, 141

Nusrat al-Din Muhammad b. Husayn b.

Kharmli, 40

Nusrat al-Din Taisi, Malik, 63n, 134, 143-4

Nusrat Khan (Malik Nusrat Jalesari), 77n, 172,195,197

Nusrat Khan, Badr al-Din Sonqur Sufi-yi Rumi, 63n, 68n, 77, 78n, 90, 112-13

Nusrat Khan, Shihab Sultani, 181, 269-71

Nusrat Shah, Nasir al-Din (rival sultan 1394-8), 309-10,319, 321

Nusrat-i Sabah, Malik, 84n, 99

Office, heredity of, 101-2, 304-5

Ogodei (Mongol qaghan), 103-5, 107, 110, 217, 227n,312

Olberli, 57, 63, 64, 76

Oqotur (Mongol general), 105, 108

Or Khan-i Ruknl, Begtemur, 90

Orus (Mongol general), 231

Oxus, R., 217, 223-4, 226, 228-30, 232

Ozbeg-bei, Jahan-Pahlawan, 34, 36 peaks, 62, 76, 125, 157, 213, 215, 280, 284

Palam, 243, 323

Palwal, 96, 128

Pandyas, dynasty, 161, 194, 207, 211; kings:

Sundara Pandya, 206-8; VTra Pandya, 206, 209, 214; see also Macbar Panipat, 310, 324

Panjab, 3-4, 6, 10-12, 21, 30, 32, 38, 106, 116, 121, 221-2, 227, 236-7, 266, 268, 300, 307, 323-4

Paramaras, dynasty, 133, 146, 198-9; kings: Devapala, 146n; 'Mahlak Deo', 198

pargana, 249 Parwan, 227

Parwans, 177, 178n, 179, 280

Pashai, 115, 234n

*Pashaitai (Chaghadayid prince), 234

*pashib, 215-16 Patan, see Nahrwala

Patiyali, 135, 321; called 'Mu'minpur', 135 *Patlahi, 302

Pattan (in Ma'bar), 207-8 Payal, 98 Persia, 5, 7, 24, 33, 49, 103, 108-9, 111, 114-15, 119, 121,

176, 184, 188, 206-7, 219, 226, 233, 237, 239, 314; see also Ilkhanate

Persian Gulf, 184-5, 193, 208, 214, 252

Peshawar, 10, 27, 30, 32, 34, 230-2

Petlad, 196

Philip IV, king of France, 220n

pilkhana, 302, 307

Pir Muhammad, 311n, 313

Polo, Marco, 115, 119, 205-6, 213, 252, 328

postal relay system, 210

Pratiharas, dynasty, 143, 144, 199; king:

'Mangal Deo', 143 prices, control of, 170, 245-7, 250; fluctuation in, 316

Prithviraja III, see Ranthanbor

*Qabtagha, 185-6

Qabul 'Khalifat!', Malik (Malik Kabir), 183, 185n,186

Qabul 'Qur'an-khwan', Malik, 187, 307, 311

Qabul Toraband, Malik, 187

Qabul Ulughkhani, Malik (shihna-yi manda), 179,245 qadi, 25, 281

QadiJalal, 274-5 qddi-yi lashgar, 25

Qadr Khan, Husam al-Din Pindar Khalji, 189-90, 261, 267

Qaidu (Mongol khan), 110, 217-21, 227, 235-6

qalandars, 164

Qandahar, 121,218

Qangli, 42, 63

Qarachil, 129, 240, 298, 307; invasion of, in Muhammad b. Tughluq's reign, 152n, 164-5,261,264-

5,269

Qarachomaq, 76

Qara-Khitan (or -Khitai), 5-6, 11, 13, 63, 65, 326

Qaraqush Khan, Ikhtiyar al-Din Aytegin, 63n, 69-70, 88, 96, 97, 105, 127

Qara'unas, 122, 178, 235, 311, 328; see also Neguderis Qashani, 103 Qasur, 221

Qaymaz-i Rumi, 28 9

Qays, 184,208

Qaysar-i Rumi, 168, 271

Qazaghan (Mongol amir), 235, 311-12

Qilich Khan, Jalal al-Din Mas'ud b. Jam, 87, 90-3, 101,327

Qinnawj (Kanauj), 6-7, 9, 12, 73n, 74n, 92, 134-6, 143, 144n, 265, 270, 307, 309, 321

Qipchaq, 57n, 63, 67n, 76

Qirabeg, Malik, see Ahmad-i *Chhitam

Qiran, Malik, 176

Qiran Safdar Malik, 184 qorabeg, 101

Qubacha, Nasir al-Din, 29, 32-5, 38-41, 104, 126, 159

Qubilai (Mongol qaghan), 108-10, 217, 236

Quhistan, 225

Qunduz, 105 quriltai, 103

Qusdar, 186

Qutb al-Din Bakhtiyar KakT, Chishtl shaykh, 159

Qutb al-Din Hasan b. 'All Ghuri, 40, 42, 67-9,72,74-5,92, 132

Qutb al-Din Mubarak Shah (sultan 1316-20), 58, 139, 189-91, 225n, 231, 249-50, 279-80; his

reign, 157-9, 160, 177; warfare against Hindu powers, 202-3, 207

Qutb al-Din Munawwar, Shaykh, 159

Qutlugh (amir-i shikar), 179

Qutlugh Khan (nephew of Sultan cAla' al-Din), 171,173

Qutlugh Khan (Shamsi amir), 49, 72-5, 90 92,97,100,125,130,140,327

Qutlugh Khan, Qiwam al-Din (governor of Deccan), 181, 213, 268, 270, 273

Qutlugh Khan, Sayf al-Din Aybeg-i Tutuq 67,78n

Qutlugh Qocha (Chaghadayid prince), 172, 197, 218-25, 228, 230, 235-6, 239, 314

Radiyya bint Iltutmish (sultan 1236-40), 46-7, 55-8, 65, 68, 87-8, 91, 97, 113, 127-8, 144-5, 278n;

her reign, 47, 67

Rahab (Ramganga), R., 136, 138

ral(rdja), 9, 19

ra'is, 246

Rajab, 182, 229n, 248

Rajasthan, 54, 123, 156, 173, 194

Rajput, 9

Ral(Rarh), 141-2

Ramadeva, Yadava king of Deogir, 176, 195, 201-2,208-9,214

rdna (ranaka), 9, 19, 21

*Randhaval, 179

Ranthanbor, 10, 30, 39-40, 57, 86n, 171, 173, 174, 194, 198, 215, 221n, 253; Muslim attacks on,

123-4, 128, 132-3, 172; conquest of, 197; Chawhan kings of: Hammiradeva, 130-1, 197, 215; Prthviraja

111,9-11, 16,42, 130; his son, 19; Vagbhata ('Bahar Deo'), 132; Valhanadeva, 132

Rapri, 135, 175, 191,200,302

Rashid al-Din Fadi-Allah, 49, 103, 225n; his correspondence, 154

Ratan, 186,271

Ravi, R., 113, 117,228

rawat (rauta), 9, 125, 214, 242, 248-9

Rayhan, Imad al-Din, 72-4, 79, 93, 100-1

Rayy, 4

Rewa, 308

Rewari, 71, 128-9,243

Rohtak, 310

Rudradeva II, Kakatiya king of Arangal, 194, 204-5, 208-9, 214-15, 285

Rukn al-Din, Suhrawardi shaykh, 160, 328

Rukn al-Din Hamza, 41 -2

Rukn al-Din Kayka'us (sultan of Bengal), 82n, 95, 142

Rukn al-Din al-Malati, Egyptian shaykh, 272,296

Rum, 112, 179; Rumis, 62, 189

Rupar, 117

Rurki, 134

Rus, Russia, 179,315

Rustam-i Yahya, Malik, 190

Sa'adat Khan, 309

Sachor, 195-6, 198

Sadharana, 279

sadi (unit of a hundred villages), 249

Sadr al-Din 'Arif, 190-1

Sadr al-Mulk Najm al-Din Abu Bakr (wazir), 71-2

al-Safadi, 154 Sagar, 211,256-7

al-Saghani, Radi' al-Din al-Hasan, 37-8,44-5 sahs (Hindu bankers), 241, 279 Sakit (Sekit), 135

SalaDin, 20 Salar Mas'ud, 139

Salar, 'Izz al-Din Muhammad, 42, 68i

Sali Noyan (Mongol general), 108, 111, 113-14, 116-17, 122,236-8

al-Salih Ayyub, Egyptian sultan, 76 Salt Range, see Kuh-i Jud Samana, 24, 77, 82, 117-18, 127, 166, 175, 178, 187, 190, 223, 227-9, 231-2, 243, 306-7,309-10,313

Samara Singh, 186

Samarqand, 63, 116n, 313, 320

Sambhal, 136, 138,188,288

Sambhar (Sakambhari), 9;

Sambhar Namak, 130

Sammas, 300-1; rulers (Jams): 'Ala' al-Din Jawna, 300; Banbhina, 188, 300, 311; Tamachi, 300

SandTla, 200

Sanjar (Seljuk sultan), 5, 12

Santur, 73, 86n, 125,133n;

Ranpal, raja of, 129

sar-i jandar, 25

sar-i silahdar, 25

Saral-yi 'Adi, 245-6, 250

Sarang Khan, 310-11, 313, 318

Sarban (Mongol prince), 217-19, 221n, 224

Sargadwari, 266, 269-71, 273n Sarju, R., 53, 125

SarsatT, 30, 35, 100, 130, 179, 192, 313

SartTz, 'Imad al-Mulk, 183, 190, 271, 273, 276

Sariir al-Sudur, 48

Satara, 212

Satga'un, 142

Sayf al-Din Aybeg (muqta' of Uchch), 88, 97, 105

Sayf al-Din Aybeg ShamsT-yi 'Ajami (dadbeg; later 'Adil Khan), 63, 73n, 77, 96, 97-8

Sayf al-Din Aybeg-i Yaghantut, 63n, l0ln, 141

Sayf al-Din Firuz, 42 Sayf al-Din Ikit Khan Aybeg-i Khita'i, 63n, 73n,74n,106

Sayf-i Shamsi (father of Amir Khusraw), 95 Sayfi, 103, 11 In

sayyids, 18, 41, 58, 140, 191-2, 233, 292, 316

Sayyids, dynasty, 170n, 233, 319, 321-3

Seljuks,4-6, 16-17,24, 326

Senas, dynasty, 9, 13, 19, 141

Shabankara, 184

Shabankara'i, 155

Shaburghan, 218n, 225

Shadi, Malik, 177

Shadi, Malik (son-in-law of Tughluq Shah I), 179

ShadI Khan (son of Sultan 'Ala' al-Din), 157, 176, 202n

Shafurqani, 'Imad al-Din (grand qadi), 101

Shah Rukh, 322

Shah Terken, 46, 59

Shahik (Azhdar Khan), 81, 82n, 118

Shahin, Malik, 175, 198

Shahu Lodi, 183n, 268, 271 n, 274

Shal (Quetta), 115

Shams al-Din (grand qadi), 73

Shams al-Din Dawlat Shah (sultan of Bengal), 95

Shams al-Din Firuz Shah (sultan of Bengal), 95, 142, 200

Shams Khan Awhadi, 318

Shamsids, dynasty, 44, 46, 56-7, 60, 76, 78, 99, 174,215,333

Shamsis (slaves of Iltutmish), 43, 48, 55, 61-70, 77-8, 88, 96; divisions among, 56, 70-6

Shankara, 301

Shansabanids, see Ghurids SharafQa'ini, 173, 176,243

Shari'a, 25, 156, 169, 233, 266, 281

Shihab al-Din Jami,

Shaykhzada, 160

Shihab al-Din Mas'ud KhaljT, 80, 83

Shihab al-Din cUmar (sultan 1316), 157, 176-7

shihna (intendant of the khalisa, governor), 96; (Mongol resident), 103, 105, 112 shihna-yi p'd,

68n

Shir (Shira; Mongol general), 231

Shir Khan, Mahmud Beg (muqtac of Samana;d. before 1364), 166, 190; his sons, 190

Shir Khan, Nusrat al-Din Sanjar, 63n, 66n, 71-2, 73n, 74-5, 77, 89-90, 94n, 97, 99, 106, 111-13,

116-17, 127-8, 134, 144 Shir Shah, 324

Sidi Muwallih, 54, 57, 83, 101, 160

siege warfare, 215-16; see also maghribi; pashib

Sikandar Khan, 307

Sikandar Shah (sultan 1489-1517), 323-4

Sikandar Shah (sultan of Bengal), 299-300, 302

Sikandar Shah, 'Ala' al-Din (sultan 1394; formerly Humayun Khan), 309 silver, supply of, 261-2,

302

Sind, 7, 30, 38, 53, 80, 86, 88, 94, 100, 111, 113,116, 117,119, 126, 166, 168, 183n,186, 221, 229,

238, 270, 286; early Muslim rule in, 3, 6, 15, 20, 282-3; see also Sammas, Sumras

Siraj-iTaqi, 184n, 208

Siraji (poet), 45, 138

Sirat-i Firuz-Shahi, 155, 168,282

Sirhindi, Yahya b. Ahmad, 46, 50, 151, 156, 162, 169-70, 322

Siri, 59, 189, 222, 232n, 259-60, 310, 321

Sirmur, 129, 169, 23 5n

Sirpur,202,213

Sistan, 40, 80n, 105, 112, 121, 186, 220, 226; in narrower sense, of Ghur and Gharchistan, 217

Siwalik,29, 100,130,227-8

Siwana (Sevana), 198, 215, 253

Siwistan (Sehvan), 33, 106, 127, 168, 178, 219-20,223,271,303

Siyalkot, 36, 113

Sodra, 36, 111, 113

*Sogedei (Mongol chief), 219-20, 236

Somnath, 6, 195

Sonpat, 310

Srirangam, 207

sufis, 154, 159-61, 164; warrior-, 210-11; sufi shaykhs, 159-61, 164, 233, 293; as kingmakers,

160; see also Chishtiyya, Suhrawardiyya

Suhrawardiyya, sufi order, 159, 161

sukunat-ghari (or -garhi; dwelling-tax), 243, 248, 263

Sulayman range, 11

sultani (sobriquet of a slave of reigning sultan), 62, 174

Sultankot, 96, 143

Sumer (rai of Etawa), 188, 302, 308-9, 321

Sumras, 8, 35n, 126-7, 270-1, 276, 298, 300-1; rulers: see Chanisar, Hammir Duda

Sunarga'un, 141-2, 182, 200, 257, 299

Sunbul, Ikhtiyar al-Din, 179

Sunnam, 74n, 77, 127, 175, 190, 227, 243

Sutlej, R., 33, 117-18,221

Sylhet (Sirihat), 142,211

Syria, 17, 20, 44, 63, 80, 110, 112, 114, 115, 219,266

Tabarhindh, 10-11, 26, 30, 35, 47, 67, 72, 74, 89,92,96-7, 111-13, 127 Tabaristan, 112

Taghai, 184, 197, 270, 275-6, 298

Taibu (Mongol general), 228, 328

Taj al-Din Abu Bakr b. Ayaz, 89

Taj al-Din 'Iraqi, 101

Taj al-Din Ja'far, 191

Taj al-Din b. Qutlugh Khan, 78n

Taj al-Din Sanjar-i Qabaqulaq, 63n, 69, 99, 136

Taj al-Din Sanjar Siwistani, 74, 93

Taj al-Din Turk, Malik, 188, 191, 303

Tajiks, 18, 25, 39, 41, 61, 65-8, 75, 79, 179

Taliqan, 105, 218n talwandi (or talwara), 127, 237 tama, 107-8

Tana, 204 tanga, 37

Taqi al-Din Abd al-Rahman, 208

Taraghai (Mongol general), 220, 222-4, 227-8, 235, 244, 253 Tara'in, 67, 69, 70; battle at

(1188/91), 10, 12; battle at (1192), 7, 10-11, 16-18, 27; battle at (1216), 30

*Tartaq (Mongol general), 174, 227-8, 230

Tatar Khan/Tatar Malik (adopted son of

Tughluq Shah I), 166, 182, 200, 302 Tatar Khan, Muhammad (ruler of akhnawti), 93-4

Tatar Khan b. Wajih al-Mulk, 309-10, 319 taxation, 173, 197, 200, 242-4, 248, 250, 262-3;

collection, 99-100, 269; figures, 251-2, 315-16; tax-farming, 269-70, 273; uncanonical taxes, 168, 262-3,

266, 272, 316; see also chard'i; jizya; kharaj; sukunat-ghari

Tegin, Ikhtiyar al-Din (muqtac of Awadh), 174, 179-80, 182

Teguder (Mongol prince), 115n

Temuge (Mongol prince), 107

Temur (Mongol leader), 117, 185

Temur (Mongol qaghan), 220

Temur (son of Abachi, Neguderi chief), 225-6

Temiir (son of Ebugen, Mongol prince), 218, 224

Temur (Tamerlane), xiii, 103, 152, 159, 170, 227, 232, 311-14, 318-20; sources on, 156

Temiir, Ikhtiyar al-Din (muqta' of Chanden), 174, 179-8O

Temiir Buqa (Mongol prince), 222

Temur Khan, Qamar al-Din Qiran, 63n, 91, 97, l0ln, 140, 144-5 Temiir Khan, Sonqur-i cAjami,

66n, 77

*Teniz Khan, Taj al-Din, 71-2, 74, 90, 138 Terdol, 212 thakurs (thakkuras), 9, 21

Thanesar, 232, 287

Thangir, 10, 27

Than (Thar), 220, 236

Thatta, 127, 162, 168-9, 276, 314, 317; see also Sammas Tibet, 187

Tiginabad, 121,220,225

Tilang (Telingana), 161-2, 185-6, 194, 196, 201, 204, 209-10, 212, 223, 282, 293; reduction of,

204-5; loss of, 268

Tilpat, 173

Tilsanda, 136

Timurids, 170n, 322

Tipperah, 141

Tirhut, 86n, 125, 139-40, 172, 201; alias

Tughluqpur, 201

Tirmid, 63, 185,232

Tirmidi, Ghiyath al-Din, 185;

Khudawandzada Qiwam al-Din (later Khudawand Khan), 185-6, 305n; Sayf al-Mulk, 186

Todars, 127

Toghan Khan, cIzz al-Din Toghril, 63n, 91-3, l0ln,140, 142

Toghril (rebel amir under Balaban), 55, 78, 94, 124-5, 141,239

Toghril Khan, Ikhtiyar al-Din Yuzbeg, 63n, 73n, 74n, 89, 91-2, 94, 97, 111n, 141-2, 290

Tolui, 107; Toluids, his descendants, 109, 226

Tomaras, 9, 188, 321, 323; see also Govindaraja, Gwaliyor

Transcaucasia, 110

Transoxiana (Ma wara' al-Nahr), 5, 41, 90, 104, 109, 112, 116, 119, 122, 162n, 185, 229, 233-4,264,266,311-13

Tribeni, 142

Tughluq Shah I, Ghiyath al-Din (sultan 1320-4), 56, 98, 140, 188-92, 199, 210, 256; lost manaqib

of, by cAfif, 152; his origins, 178; as Ghazi Malik (prior to his accession), 127, 142, 157-8, 248; a veteran

of the Mongol frontier, 220, 227-9, 231; his coup d'etat, 178-9; his reign, 157, 160-1, 179-82, 250-1; date of

his death, 330-1; conquests of, 200-1; his sons, 182

Tughluq Shah II, Ghiyath al-Din (sultan 1388-9), 303, 305-9, 318

Tughluqabad, 259-60

Tughluqids, dynasty, 153-5, 161 and passim, 318,335

Tughluqpur, 188, 302, 308

Tukharistan, 105, 112

*Tulabugha Bughda, 179; *Tulabugha Nagawri, 179

Tulak, 11,25

Turghai, 81

Turkestan, 5, 109, 112

*Turki, 81

Turkish slaves, 4, 6, 11, 12, 31, 6l-84passim, 174, 183-4, 326; supply of, 64, 174; see also

Ghiyathis, Shamsis

Turkmens, 82n, 183n, 322, 326

Turks, 4, 17-18, 25, 29, 39, 41-3, 57, 61, 64-5, 67-9, 72, 75, 82, 174, 183-4, 213, 219, 278, 326;

term includes Mongols, 113, 232, 328; see also Turkish slaves, Turushkas

Turumtai, 81

Turushkas, 123, 130, 133, 143

Ubayd(-i Hakim), 180 Uchch, 27, 30, 33, 35, 39-40, 49, 71-2, 75, 89, 92, 96-8, 105-6, 111-13,

116-17, 168, 178,313

Uddandapuri, 140

Uddharan, 188, 302, 308-9

Uighuristan, 236

Ujjain, 19n, 146, 179, 199, 251, 329

Ulaghchi, 81, 82n ulama' 18, 41, 47-8, 58, 164, 168, 233, 258, 272,316 Ulugh Khan (Almas Beg),

171-2, 175, 195-7, 204, 221, 245n; see also Balaban, Muhammad Shah b. Tughluq

Ulughkhani (Hajji al-Dabir), 151 ulus, 107 cUmar, Malik, 306

cUmar Khan (nephew of Sultan 'Ala' al-Din), 173

*Umardan, 142 al-

cUmart, 154, 163, 165

Urgench, 253

Vaghelas, dynasty, 195-6; king: Karnadeva, 173-4, 195-6,215

Venetians, 253

Vijayanagara, 162, 204, 268, 298; influence of Delhi Sultanate at court of, 210

Wahid al-Din Qurayshi, 177, 190

Wairagarh (Basiragarh?), 202n

wajhdars, 316

Wajih al-Mulk, see Zafar Khan

wakil-i dar, 42

Wassaf, 49, 103, 154

wazir, office of, 25, 96, 187

Yadavas, dynasty, 87, 147n, 173, 175, 194-5, 196n, 201-3, 211; kings: Melugideva, 203;

Singhanadeva, 202; see also Ramadeva

Yaklakhi, Malik, 243

Yaklakhi, Malik (muqta' of Samana), 178, 179n

Yamuna, R., 58, 86, 134-5, 145, 191, 223, 232,245,251,313

Ya'qub, nazir, 245-6

Yaqut, Jamal al-Din, 67-8

Yasa'ur (Chaghadayid prince), 226, 234

Yasa'ur (Mongol general), 217

Yavanas, 123, 142

Yazdagird, 164

Yemek (Kimek), 57n

Yemen, 275

Yildiz, Taj al-Din, 13, 25-6, 28, 30, 31n, 35, 39,62

Yinaltegin, Taj al-Din, 40, 42, 105

YuanShih, 104

Yughrush, 80

Yuri-nan, 261

Yusuf (na'ib of Ghazi Malik Tughluq at

Deopalpur), 179 Yusuf (the Patriarch Joseph), 63

Yusuf-i Ahl, 155; his Fara'id-i Ghiyathl, 155,233-4 Yusuf-i Bughra, 232, 276

Zabulistan (Zawulistan), 105, 112, 227, 298

Zafar Khan (I), Taj al-Din Muhammad Lur Farsi, 188,299,304 Zafar Khan (II; formerly Darya

Khan), 188, 299,304 Zafar Khan, Hizabr al-Din Yusuf, 171-2, 219,221n,222, 273 Zafar Khan Wajlh al-

Mulk (Sadharan), 188, 308-10,318-19

Zafarabad, 182n, 200, 251; see also Jawnpur Zakariyya (Mongol general), 231

zamindars, 126, 210 Zaminda war, 11 Zayn al-Din, Shaykh, 159